What do the Ice Core Bubbles Really Tell Us?

Guest Opinion: Dr. Tim Ball

At a recent public presentation titled “Whither the Weather” the most frequently asked question popped up once gain. “How do they know the temperatures from thousands of years ago, as in the Antarctic ice record.” I gave the standard answer about layers of ice, extraction of air from trapped bubbles and then comparison of the Oxygen 16/18 isotope ratio, which varies with atmospheric temperature. As always, people are dazzled by that, and even though they don’t understand, it sounds plausible. The trouble is that every time I give the answer, it triggers my long-term concerns about the nature of glaciers, glacial ice, entrapment of the bubble, and recovery methods of the air in the bubble. I spent hours discussing all aspects of glaciers and ice cores with the late Dr. Fritz Koerner, one of the few people to study glaciers in the Arctic and Antarctic.

Like all great scientists, he was very aware of the limitations of knowledge, data, and mechanisms in his area of research. I specifically recall him telling me that his work on Baffin and Ellesmere Islands was indicating that temperature changed before CO2 before it was disclosed in the Antarctic record.

The short answer to the question posed in the title to this article is virtually and practically nothing. They definitely do not tell us what is claimed, that is, accurate representation of the state of the atmosphere including temperature in individual years. This is why one of the world’s experts on atmospheric chemistry and ice cores Zbigniew Jaworowski M.D., Ph.D., D.Sc., wrote,

“It was never experimentally demonstrated that ice core records reliably represent the original atmospheric composition.”

Dr. Jaworowski was so respected as an atmospheric chemist that he was chosen to lead the United Nations (UN) investigation into the impact of the Chernobyl disaster. Of course, none of that muted the attacks on him because of his well-reasoned, fully documented views on climate change due to human production of CO2. Undoubtedly, the ad hominems will appear in comments about this article.

Just a few facts about the formation of the glaciers illustrate the problems. Glacier ice forms as precipitation fall above the snow line and accumulates in layers most of which survives the summer melt. These layers build up and change from snow, through firn (granular snow) then meld into layers of ice under the weight and pressure of overlying layers (Figure 1). This process of converting snow to ice takes years and varies depending on a variety of factors but especially temperature. The question is which year does the final bubble represent. How does it remain isolated and insulated from contamination in a very wet, dirty, and constantly changing situation? The answer is it doesn’t, and there is no way of saying that any layer at any level represents a particular year or even a span of years. As I recall Koerner told me that a core sample of eight meters was required at the bottom of Antarctic ice to yield enough sample for a single reading. The problem is at those depths, eight meters of ice represents 10,000 years of compression. How is that useful for climate when a single sample for the entire period of the Holocene Optimum.


Figure 1

A close-up of the layers of one glacier shows the problems of dirt and lack of distinction that occurs even above the glacial ice (Figure 2).


Figure 2.

At an approximate depth of 50 m (150ft), the ice becomes plastic and displays different characteristics than in the Brittle layer. This is why crevasses, that is cracks in the ice surface, only extend down to the Plastic Layer. In that Layer, ice flow deforms and gradually, with depth, most of the gases in the ice are squeezed out. Fischer summarizes the major problems with ice core bubbles as follows,

Due to glacier flow and in the top 50-100 m due to firnification annual layers in the ice become thinner with depth (Fig. 3). This restricts the resolution of ice core parameters in deeper ice and makes dating of ice cores (a crucial prerequisite for the interpretation of climate records) a difficult task.

However, going deeper into the ice sheet, where the hydrostatic pressure is increased, the air bubbles become smaller and smaller due to further deformation (creep) of the ice and the density increases slowly until all bubbles disappear.

The ice at the depth of the bubble enclosure is older than the enclosed air. This ice age/air age difference Δage has to be taken into account when comparing e.g. greenhouse gas concentration and temperature records from the same ice core.

The bubbles at a certain depth are not occluded at the same time. This implies that in a given sample the age of the air in individual bubbles is different. In addition, air needs a few years to diffuse down to the depth of bubble enclosure, also leading to a secondary broadening of the age distribution of the air at a certain depth.

So, the age of the ice is not the same age as the ice that surrounds it, and even the age of the air between bubbles is different. Fischer notes,

The gas records allow only a resolution of decades to a few centuries because of the slow bubble enclosure process (see below).

Apparently, they believe most of these differences are filtered out with statistics, but that doesn’t cover the useless nature of the results. To mask the problems even more, they apply a 70-year smoothing average to the raw data.

Then there is the constant flow of water across and through every portion of the glacier. Not only are glaciers dirty, as you can see in Figure 2, but they are also very wet. Water flows over, through and under them, in small amounts but also in great torrents. Every summer, even above the snow line, there is a period of melt and the water filters down through the snow in all its forms. This water constantly contaminates any bubbles within the ice, so it is virtually impossible for the air in that bubble to be uncontaminated. As Dr. Jaworowski observed,

The basic assumption behind the CO2 glaciology is a tacit view that air inclusions in ice are a closed system, which permanently preserves the original chemical and isotopic composition of gas, and thus that the inclusions are a suitable matrix for reliable reconstruction of the pre-industrial and ancient atmosphere. This assumption is in conflict with ample evidence from numerous earlier CO2 studies, indicating the opposite (see review in Jaworowski et al. 1992b).

He adds that additional assumptions are equally invalid.

1. No liquid phase occurs in the ice at a mean annual temperature of −24°C or less (Berner et al. 1977, Friedli et al. 1986, Raynaud and Barnola 1985).

2. The entrapment of air in ice is a mechanical process with no differentiation of gas components (Oeschger et al. 1985).

3. The original atmospheric air composition in the gas inclusions is preserved indefinitely (Oeschger et al. 1985).

4. The age of gases in the air bubbles is much younger than the age of the ice in which they are entrapped (Oeschger et al. 1985), the age difference ranging from several tens to several ten-thousands of years.

More than a decade ago, it was demonstrated that these four basic assumptions are invalid, that the ice cores cannot be regarded as a closed system, and that low pre-industrial concentrations of CO2, and of other trace greenhouse gases, are an artifact, caused by more than 20 physical-chemical processes operating in situ in the polar snow and ice, and in the ice cores. Drilling the cores is a brutal and polluting procedure, drastically disturbing the ice samples (Jaworowski 1994a, Jaworowski et al. 1990, Jaworowski et al. 1992a, and Jaworowski et al. 1992b).

It is interesting to note parallels in bad science between the ice core procedures and the computer models. They are based on false assumptions, inadequate untested data and are not validated. Sadly, it is a common theme of the deception that is human-caused global warming (AGW). I urge everybody to read Jaworowski’s article completely because it ties the ice core debacle into the wider debacle of international climate science.

I recall when the French scientists led by Petit, Jouzel, et al., announced the reconstruction of temperature, CO2, and Deuterium levels based on ice core data. One of them, as I recall Jouzel, warned about rushing to judgment. It was approximately five years later that research showed that temperature changes preceded CO2 changes, not as assumed, yet that is the theme still generally pushed to the public.

In the ice cores, the isotopically determined temperature signal and the signal of CO2 air concentrations are out of phase by hundreds to several thousands of years (Jaworowski et al. 1992b), with the temperature increases always preceding the rising CO2 levels, not the reverse (Caillon et al. 2003, Fischer et al. 1999, Idso 1988, Indermuhle et al. 2000, Monnin et al. 2001, and Mudelsee 2001).

All other measures agree with this juxtaposition regardless of the time period or length of record. But, even allowing for this, we have the problem that Jaworowski, nor anyone else to my satisfaction can answer.

Only recently, many years after the ice-based edifice of anthropogenic warming had reached a skyscraper height, did glaciologists start to study the fractionation of gases in snow and ice (for example, Killawee et al. 1998), and the structure of snow and firn which might play a first-order role in changing gas chemistry and isotopic profiles in the ice sheets (Albert 2004, Leeman and Albert 2002, and Severinghaus et al. 2001). Recently, Brooks Hurd, a high-purity-gas analyst, confirmed the previous criticism of ice core CO2 studies. He noted that the Knudsen diffusion effect, combined with inward diffusion, is depleting CO2 in ice cores exposed to drastic pressure changes (up to 320 bars—more than 300 times normal atmospheric pressure), and that it minimizes variations and reduces the maximums (Hurd 2006).

This is illustrated by comparing for the same time period, about 7,000 to 8,000 years before the present, two types of proxy estimates of CO2. The ice core data from the Taylor Dome, Antarctica, which are used to reconstruct the IPCC’s official historical record, feature an almost completely flat time trend and range, 260 to 264 ppmv (Indermuhle et al. 1999). On the other hand, fossil leaf stomata indices2 show CO2 concentrations ranging widely by more than 50 ppmv, between 270 and 326 ppmv ( Wagner et al. 2002).

The stomata record Jaworowski refers to is shown in Figure 3, shown for clarification with the original caption.


Figure 3

The range of variability of the stomata fits the record for the 90,000 19th century direct atmospheric readings studied by Beck. It also fits the pre-manipulated record at Mauna Loa.

Jaworowski is saying that, like dendrochronology was misused for dendroclimatology, glaciology and specifically ice core research was coopted by the global warming hysteria created by the Intergovernmental Panel on Climate Change (IPCC). Ice cores were manipulated and massaged using inadequate assumptions, lack of understanding of physical and chemical mechanisms, and masked by statistics to produce a result.

This difference (between stomata and cores) strongly suggests that ice cores are not a proper matrix for reconstruction of the chemical composition of the ancient atmosphere.

Like so many of these claims of scientific certainty about past climate reconstructions, the ice core bubbles claims bear little scrutiny. They confirm A. N. Whitehead’s warning;

“There is no more common error than to assume that, because prolonged and accurate mathematical calculations have been made, the application of the result to some fact of nature is absolutely certain.”

5 3 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
michael hart
January 20, 2018 11:02 am

As a grad student, I recall studying methods of peptide synthesis where the rate-limiting step was how quickly the aqueous solutions could be de-gassed of the carbon dioxide by-product. Simply assuming a thermodynamic equilibrium should be reached does not mean that it will be reached in any given time span.

To ever assume that the chemistry of CO2 in a water/ice environment is independent of pressure strikes me as one of the silliest assumptions I have yet come across in climate science.

Reply to  michael hart
January 20, 2018 11:48 am

Actually, they’re not really looking at the CO₂ fraction that much: of all the things well known to have nonlinear solubility of atmospheric gasses and water ice, ice + CO₂ is huge. It only mildly resists going into solid-state solution in surrounding ice under pressure. Oxygen is more resistant (but less than nitrogen, and both of those aren’t all that resistant either).

So I agree.

Richard G.
Reply to  GoatGuy
January 22, 2018 9:45 am

To add a pinch more uncertainty to the witches brew of conjecture that is proxy evidence, consider the phase change diagram for CO2: gas to solid (precipitate) at -78Deg C at one atmosphere pressure. This presents the possibility of there being a CO2 fraction within the snow that falls under extreme cold conditions.


“The lowest natural temperature ever directly recorded at ground level on Earth is −89.2 °C (−128.6 °F; 184.0 K) at the Soviet Vostok Station in Antarctica on July 21, 1983 by ground measurements.
A 2009 study estimated that under exceptional climate conditions similar to those recorded at Vostok 1983, temperatures higher on the plateau around Dome Argus could drop lower than −95 °C (−139 °F; 178 K).[1] On August 10, 2010, satellite observations measured a surface temperature of −93.2 °C (−135.8 °F; 180.0 K) at 81.8°S 59.3°E, along a ridge between Dome Argus and Dome Fuji…” https://en.wikipedia.org/wiki/Lowest_temperature_recorded_on_Earth#cite_note-2

Reply to  GoatGuy
January 22, 2018 11:16 am

Richard G.,

The partial pressure of CO2 is 0.0004 bar in the atmosphere. With 1 bar it may precipitate at -78ºC, but you need much lower temperatures to condense any CO2 at 0.0004 bar pressure…
Neither does 1% water vapor in the atmosphere precipitate at 100ºC, but it does if it is 100% water at 1 bar…

Reply to  GoatGuy
January 23, 2018 2:58 am

Richard G. January 22, 2018 at 9:45 am
To add a pinch more uncertainty to the witches brew of conjecture that is proxy evidence, consider the phase change diagram for CO2: gas to solid (precipitate) at -78Deg C at one atmosphere pressure.

Fortunately the pCO2 pressure in our atmosphere is much less than one atmosphere so this can’t occur. This is the same fundamental misunderstanding of the phase diagram that Steve Goddard made here.

Reply to  michael hart
January 20, 2018 11:52 am


CO2 does not reside in the ice matrix, but may “hide” in the thin water-like layer at the edge of the ice-air border. That doesn’t play any role at measurement time, as that is done under vacuum, effectively removing all CO2 and most water from the grated bubbles. That passes a cold (-70ºC) trap where water vapor freezes out and has little to no CO2 retainment (as one passes a calibration gas first with about the expected composition).

Reply to  Ferdinand Engelbeen
January 21, 2018 7:12 pm

Surely absorption of CO2 to the interface is temperature dependent and the CO2 concentration in the bubble dependent on the temperature difference after removing the core and bubble size.

Reply to  michael hart
January 20, 2018 7:35 pm

This post does not make clear that the 18O/16O ratios from which temperature of the ambient air is calculated are independent from air bubbles. Bubbles are used to measure entrapped CO2. As 18O/16O records temperature at time of snowfall, whereas bubbles are not formed or closed for some time afterward, the two data sets do not measure the same time.
Besides 18O/16O in ice, a few studies have used measured isotopic fractionation in nitrogen and argon trapped in bubbles along with CO2 to calculate temperature at the time of bubble closure. This approach has larger uncertainties, but suggests that generally similar times are being measured.
The question of gas diffusion after entrapment is complicated by possible differences in diffusion rates of various gases after bubble entrapment.

Reply to  donb
January 21, 2018 3:30 am

Yes the 18O/16/O ratio is most commonly measured on ice, but not exclusively. It is measured on O2 from trapped air bubbles as well. This measurement is used for a variety of purposes, most notably as a proxy for changes in global sea level, given that O2 is produced predominantly by plankton and so supposedly directly reflects the O18/O16 ratio in the seawater.

Reply to  donb
January 21, 2018 6:43 am

In addition, δ18O in N2O is used as proxy for global ice sheet volume. I don’t know the rationale behind that…

Reply to  michael hart
January 21, 2018 9:44 am

Exactly. Lab conditions are nothing like the real world actual conditions. In chemistry, conditions are everything ;).

Retired Engineer John
January 20, 2018 11:08 am

We have carbon dioxide levels and temperatures for about 75 years; why can’t these be used to calibrate the readings we get from ice cores?

Reply to  Retired Engineer John
January 20, 2018 11:39 am


The problem is that there is little overlap between ice core air and ambient air: it takes at least 40 years to close the bubbles in the ice and the most recent ice thus is from the end ’70s. And accurate CO2 measurements at the South Pole started about 1958. That gives only a 20 year overlap (1958-1978), nicely within the accuracy (+/- 1.2 ppmv – 1 sigma) of the ice core measurements:


Michael Jankowski
Reply to  Ferdinand Engelbeen
January 20, 2018 12:48 pm

Well it raises some red flags when I see that “…ice core results have been corrected…”

Makes me wonder how much the corrections were made to suit the air record.

Reply to  Ferdinand Engelbeen
January 20, 2018 1:36 pm


Indeed, if one sees the word “corrections”, be aware… In this case the “corrections” are less than 1% of the measurements and are proportional to the problem: in stagnant air, which is the case for many years in firn, the heavier molecules and isotopes tend to increase at the base of the air column, just before closing the bubbles. That is corrected for by measuring the increase in 15N/14N ratio, which seems to be practically unchanged in the atmosphere over the past millions of years. Based on the increase in ratio of δ15N, the enrichment of CO2 in air due to gravity is corrected.

M Courtney
Reply to  Retired Engineer John
January 20, 2018 11:46 am

Because it takes about 50 years for the snow to become firn and then ice.
And even then it doesn’t tell us anything about what happens to the ice over the years.

For many years on this blog I have pointed out that ice melts and reforms all the time at surfaces due to vibrations (the wind, for example). Any air bubble concentration will be smoothed eventually. It may take a decade or two before that happens, but we are talking about centuries here.

Brian Hatch
Reply to  Retired Engineer John
January 20, 2018 2:17 pm

We have CO2 records using the Pettenkoffer method from 1822 to 1960. CO2 was not a stable level, with high recordings in the 1870s and 1940s. The same method should be used now to compare what levels are recorded at Manua Loa and elsewhere.

Reply to  Brian Hatch
January 20, 2018 3:23 pm


If used with fresh reagentia and good skill, the old wet method was quite accurate (+/- 10 ppmv). The problem was not the accuracy, the problem was where was measured: in towns, forests, under, in between and above growing crops,…

Take the longest series compiled by the late Ernst Beck with a 80 ppmv “peak” around 1942 (not seen in any other proxy or ice cores). That “peak” was mainly from two places: Poonah, India and Giessen, Germany. The first was measuring CO2 around growing crops. Completely worthless for “background” CO2. The second measured CO2 three times a day at different heights at the edge of a small town and forest. Average SD overthe full period 68 ppmv (one sigma). A modern CO2 monitoring station is now at a few km from the original place: average SD over long term data some 35 ppmv. Mauna Loa: 4 ppmv. South Pole: 1 ppmv.
Here the differences for several stations (Mauna Loa, Barrow, South Pole) compared to Linden/Giessen over a few days without wind and under inversion, which makes things worse(all raw hourly data):


Only CO2 levels of that time measured on seaships or coastal with wind from the seaside are of value and these are all around the ice core levels for the same period. Unfortunately, there are no such data for the period 1935-1955.

Reply to  Brian Hatch
January 22, 2018 6:19 am

which means: CO2 is NOT a so well mixed gas, after all. For absorption of IR, what’s matter is not the average, it is the actual level when some radiation is to be absorbed. Up to 200ppm difference in a day.

Reply to  Brian Hatch
January 22, 2018 6:53 am


Only in the first few hundred meters over land one can’t find “background” CO2 levels (5% of the atmosphere). That is where a lot of sources and sinks are at work and mixing at low wind speed is inadequate. Once you measure above a few hundred meters (regular airplane flights), you will find the same levels as at Mauna Loa, taking into account the seasonal changes and the NH/SH barrier for air exchanges…
Here for the Rocky mountains with inversion over the valley floor:


You need a substantial change in the full 70 km CO2 column to have a measurable effect on temperature: even if you have 1000 ppmv in the first 1000 m all over the world, the effect, according to Modtran, is less than 0.1ºC.

January 20, 2018 11:12 am

One good summer could melt thousands of years

Can someone elaborate on this…..” pre-manipulated record at Mauna Loa.”…..

Jim Masterson
Reply to  Latitude
January 20, 2018 11:31 am

One good summer could melt thousands of years

Which is why Greenland ice cores are usually not considered to be reliable. Antarctic ice cores are not supposed to suffer from this problem. I remember discussing ice cores ten years ago, and we usually agreed that Greenland ice cores were suspect. Today, I’m surprised when someone brings up Greenland ice cores. What’s changed?


Reply to  Jim Masterson
January 20, 2018 11:46 am


Greenland ice cores are not reliable for CO2 levels due to interaction from sea salts (carbonates) with frequent highly acid dust from nearby Icelandic volcanoes.
Summer melt happens about once in 100 years at the summit and hardly affects the rest of the ice core for the temperature “proxy” (18O/16O ratio or D/H ratio).

One point to be taken into account is that the temperature proxy is partly from the water vapor “catch” area, but mostly from where the vapor freezes to snow, thus more balanced towards polar temperatures. That overestimates (semi) global or regional temperatures.

Reply to  Jim Masterson
January 20, 2018 11:47 am

Nobody uses CO2 measurements from Greenland ice-cores, since they are known to be completely unreliable. Oxygen-isotope measurements from the ice itself is much less problematic, a few centimeters of melted snow every century or so won’t make much difference there.

Reply to  Jim Masterson
January 20, 2018 12:16 pm

So that’s 10 times in a 1000 years…..how would anyone know….those years disappeared, melted away

Reply to  Jim Masterson
January 20, 2018 1:20 pm


Melt layers are only from one year or a few years at the Greenland summit (and at coastal domes in Antarctica). Even so, the melt layer(s) are easily recognised because of their density and one can estimate the number of years lost based on their thickness.

Reply to  Jim Masterson
January 20, 2018 1:55 pm


Jim Masterson
Reply to  Jim Masterson
January 20, 2018 3:10 pm

So my comment would have been “more correct” if I had said, “ONE of the reasons.” Thanks for replying guys.


Bill Treuren
Reply to  Jim Masterson
January 21, 2018 10:17 am

The frequency of the “100” year melts would be a good proxy for some sort of climate metric.
Does anyone have these easily read layer on a year frequency basis?

Reply to  Jim Masterson
January 21, 2018 11:06 am

Jim Materson,

Brand new method (2015) for you (very interesting stuff):

Melt layers are easily recognised in recent ice as bubble-free layers, but in deeper layers all ice is bubble-free due to clathrate formation. The researchers invented an elegant method by using the change in ratio’s between the noble gases Kr/Ar and Xe/Ar, which are elevated near melt layers due to their different solubility in liquid water. That shows that both the current and the previous interglacial, the Eemian, 127–118.3 kyear BP has several melt layers. That period was a lot warmer than today, at least at the Greenland summit.

No melt layers were detected at the WAIS divide (Antarctica), despite near-bubble free layers, which may have been formed by other means that melting.

Reply to  Latitude
January 20, 2018 12:26 pm

” It also fits the pre-manipulated record at Mauna Loa”

Yes, I don’t like unsubstantiated claims like that. But then I don’ t take much notice of Dr Ball’s writings which are usually just that.

Reply to  Greg
January 21, 2018 7:27 pm

Keeling wrote in the early 70s that his method could measure global CO2 levels to 1 ppm (hence focusing on that period in the above plot). The temperatures on the plot are scaled to a third so monthly temperature anomalies are ±0.03 (and 1 SD estimates). Even with perfect measurements, seeing such a correlation would have me dismissing the work as fiction.

Reply to  Greg
January 29, 2018 3:46 am


You have the wrong end of the stick again and you should join us all in thanking Dr Ball for opening up such uncertainty for discussion.

sy computing
January 20, 2018 11:16 am

“I specifically recall him telling me that his work on Baffin and Ellesmere Islands was indicating that temperature changed before CO2 before it was disclosed in the Antarctic record.”

What was this supposed to say please?

sy computing
Reply to  sy computing
January 20, 2018 11:27 am

“…temperature changes preceded CO2 changes”

maybe the above?

Reply to  sy computing
January 20, 2018 11:31 am

I think what it is saying is that the fact that temperature changed before CO2 was first discovered in the Arctic records (Baffin, Ellesmere) – before it was discovered in the Antarctic records.

sy computing
Reply to  Mike Jonas
January 20, 2018 11:39 am

thank you

Reply to  Mike Jonas
January 20, 2018 12:31 pm

That just leaves the alleged “manipulated” Mauna Loa data.

Keeling was a dedicated and serious scientist. I really don’t appreciate that kind of hit and run dismissal of his life’s work. If you Tim Ball is going to besmirch his work he’d better come up with something that is at least meaningful and verifiable. Not slander.

If he is claiming someone has manipulated it since then, he needs to say exactly what he means.

Reply to  Mike Jonas
January 20, 2018 11:06 pm

Dr. Ball didn’t use the word “manipulated”. Greg writes ‘”i don’t take much notice of Dr Ball’s writings” but then runs his mouth anyways. What an ass hat.

Reply to  Mike Jonas
January 20, 2018 11:11 pm

Turns out he did use the word manipulated. I’ll keep the ass hat on then for a while 🙂 apologies Greg.

January 20, 2018 11:30 am

Dear Dr. Ball,

While do appreciate your many years of work in the skeptic community, in this case you are completely wrong.

– The temperature proxy is derived from δD and δ18O from water in the ice, not from the enclosed air. Depending of the accumulation rate, even layers of 1 year resolution over the past 110,000 years can be recognised and measured, as is the case for the Greenland summit.
– One doesn’t need meters of ice nowadays, a few cm is sufficient with modern sublimation techniques, cryogenic separation and mass spectroscopy. For the isotopes in the ice mass that is in many cases enough for yearly resolution, for the enclosed gases, that depends of the accumulation rate: from less than a decade over the past 150 years (Law Dome) to 560 years for the past 800,000 years (Dome C).
– Ice cores are choosen at the top of domes, because that is where there is little to no sideward movement, little to no cracks, etc.

Then there is the constant flow of water across and through every portion of the glacier.

Sorry, we are NOT talking about glaciers which are largely melting in summer. Ice cores are taken at domes with average temperatures between -20ºC (coastal) and -40ºC (inland). Even at the Greenland summit, mid-summer surface melt only happens once in about 100 years. That influences the ice age – gas age differences for which is corrected. How much water do you think percolates at -40ºC in the Vostok or Dome C ice cores?

Last but not least: please let the late Dr. Jaworowski rest in peace, together with his completely outdated (1992!) ideas about ice cores. He was an expert about metals (ions) in ice cores as result of the Tsjernobyl disaster, but as far as I know, he never made one analyses of CO2 or other gases in ice cores, only wrote a lot of objections. These were completely, point by point, refuted by the work of Etheridge e.a. on the high resolution ice cores of Law Dome in 1996.
Dr. Jaworowski made several statements which were physically impossible, like CO2 escaping (preferentially?) via cracks in the ice to the outside world, while inside one found 180-280 ppmv and the outside air in the labs was at least 380 ppmv (or more in enclosed rooms with people in it).
See further my comment on his objections:

About modern ice core measurements, see some interesting background:
The original work of Etheridge from 1996:
onlinelibrary.wiley.com/doi/10.1029/95JD03410/full (if available, at this moment their web site is in maintenance?)

M Courtney
Reply to  Ferdinand Engelbeen
January 20, 2018 11:48 am

Either the ice cores are smoothed in the way Dr. Jaworowski said or the stomata data is wrong.
And there are lots of different plants used for stomata data.
Not many ice domes.

Reply to  M Courtney
January 20, 2018 12:03 pm

Of course the ice data are very smoothed. It takes decades to millenia before the snow turns into ice and there will inevitably be some diffusion, however slight, even in very cold ice.

On the other hand the stomata data are probably “de-smoothed”. Remember that they record the CO2 level around a particular tree during a particular spring when the stomata develop. And on that scale CO2 is quite
far from “well-mixed” whatever the climate-modellers may claim.

Reply to  M Courtney
January 20, 2018 12:09 pm

M Courtney,

CO2 smoothing (resolution) depends of how fast the air bubbles are closed. That depends of the local snow accumulation which is between 1.2 m ice equivalent per year at Law Dome and a few mm/year at high altitude inland domes like Vostok and Dome C.
That makes that the resolution is less than 10 years at Law Dome and up to 560 years (Dome C) and 600 years (Vostok).
The drawback of high accumulation is that rock bottom is reached at 150 years back (gas age) for Law Dome but 800,000 years for Dome C…

The point is that ice core CO2 are direct almost global CO2 measurments with the same equipment as used for CO2 measurements as in ambient air, be it smoothed.
Stomata data are proxies which are influenced by local CO2 levels over ground: stomata (index or density) is the result of the average local/regional CO2 levels in the previous growing season. Stomata data are therefore calibrated against… ice cores over the past century, but nobody can be sure what happened to the local CO2 levels over warmer and colder periods (like MWP-LIA and up to now), with extreme changes in land use in the main wind direction and even changes in main wind direction in certain periods…

Thus if you see that stomata data in average differ from the average in ice cores over periods longer than its resolution, the stomata data need to be recalibrated towards the ice cores, not the opposite…

Reply to  M Courtney
January 20, 2018 12:25 pm

M Courtney,

Indeed, there are not that many ice cores, but they are from places with extreme differences in temperature and accumulation rate, thus resolution. Even so all ice cores show the same CO2 levels (+/- 5 ppmv) for the same average gas age with overlapping periods back in time. That means that once the bubbles are fully closed there is no measurable diffusion of CO2 in the ice, or you should see large differences and a decreasing CO2/temperature ratio (of ~8 ppmv/K) over each interglacial 110,000 years back in time.


Stomata is a different story and not only changes from plant type to plant type, but also with altitude (CO2 partial pressure), moisture/drought, wind direction, land use changes…

Reply to  M Courtney
January 20, 2018 12:42 pm

I have done some reading on stomata proxies, and have also read much (not all) of what McIntyre has written about them. Conclusion, they aren’t very good proxies for CO2. Too many other things going on like altitude and precipitation. There was an early discussion of the ice core versus stomata complexities here 12/26/2010.

Mike McMillan
Reply to  M Courtney
January 20, 2018 9:50 pm

Even after closure of the bubbles, there is diffusion of CO2 through the ice layers leading to smoother profiles. The rate of diffusion is very slow, but in an ice dome you’ve got tens to hundreds of thousands of years for it to occur.

J. Ahn, et al. 2008 found a way to measure it. When you have a melt on the surface, the water sinks through the snow until it hits the freezing level, which is usually only a couple inches down. As it sinks, the liquid water absorbs CO2 from the not-yet-bubbles it passes, so it has a much higher CO2 concentration than the air. These melt-refreeze layers are easy to spot because they have no bubbles; the water filled in the spaces before it froze.

CO2 diffusion through the ice occurs at the edges of the layer, so the undiffused central part of the layer will give you the original CO2 concentration of the ice. Measuring the concentration across and past the edge of the layer gives a diffusion profile that can be converted to years. This is for a 10 year thick layer:


So a hundred thousand years from now when the apes drill ice cores to see what life was like for us, will they spot the current CO2 spike? Or more importantly, have we missed some big spikes in the past?

Reply to  M Courtney
January 21, 2018 12:39 am


The work of Ahn at:
gives the theoretical migration in a relative “warm” (-23ºC) coastal ice core at Siple Dome. That shows a broadening of the resolution at 2.74 kyr BP of maximum 10% or from about 20 to 22 years at that depth. At full depth from 20 to 40 years (at 70 kyears BP). That is all. No big deal if you want to know the average CO2 levels over thousands of years. One may miss some fast peaks (which never occured in the past 60 years…), but the average CO2 levels over 40 years periods didn’t change by migration.

For the Vostok, Dome C and other high altitude ice cores at average -40ºC, the migration speed is orders of magnitude smaller (exponential with temperature), thus even over 800,000 years not detectable.
That is confirmed by calculating the CO2/temperature ratio for each warm period over cold period every 110,000 years back in time. If there was the slightest migration, the ratio between peaks and lows would decrease over time, as the high CO2 levels would migrate further and further over the low levels. That is not the case.

Reply to  M Courtney
January 21, 2018 1:28 am


About CO2 peak detection: that indeed depends of the resolution. Ice cores repeatability is better than 1.2 ppmv (1 sigma) for samples at the same depth of the same ice core. That means that a 20 ppmv peak during 1 year would be detected in the less than a decade resolution of two of the Law Dome ice cores or a sustained increase of 2 ppmv over 10 years. The third Law Dome ice core (taken at the slope) goes back about 1,000 years with a resolution of ~20 years: that shows the MWP-LIA drop in temperature (-0.8ºC) with a drop of ~6 ppmv during some period.

The current increase is about 110 ppmv in 165 years. If that is part of a two-sided (natural) oscillation, that wouldn’t be detected in the Vostok (600 year resolution) or Dome C (560 years resolution) ice cores. If it is human made one-sided, that would give a detectable peak of about 15 ppmv, still increasing…

In between accumulating ice cores like Taylor Dome have resolutions around 40 years (no exact resolution found for Taylor Dome…) and span some 130,000 years back in time. That would give a one-sided peak or a two-sided oscillation of 110 ppmv, if the current 110 ppmv peak is sustained at least 40 years.

Reply to  M Courtney
January 21, 2018 2:27 am

“CO2 smoothing (resolution) depends of how fast the air bubbles are closed.”

That is an oversimplification. Until the snow turns to ice at a depth of about 90 meters there will be some exchange with the atmosphere, decreasing with depth. In places like Vostok or Dome C this takes millenia. Also at larger depths the bubbles disappear and the gases instead form clathrates. Bubbles reform as the pressure on the ice decreases. Since annual layers are almost microscopically thin at depth each reforming bubble will contain gasses from a number of layers.

Samuel C Cogar
Reply to  M Courtney
January 21, 2018 6:58 am

Ferdinand Engelbeen – January 20, 2018 at 12:09 pm

Stomata data are proxies which are influenced by local CO2 levels over ground: stomata (index or density) is the result of the average local/regional CO2 levels in the previous growing season.

Ferdinand, your persistence in repeating the above silly arsed claim that “last year’s climatic status (Sunshine, temperature, precipitation and CO2) determines the next year’s plant growth” ….. makes me wonder why anyone would believe very much of anything that you say or write.

So, cease with your touting of “junk science” about stomata growth in hopes that it will prove/justify a few of your other silly beliefs. Tout the biological facts, …… to wit:


Stomatal density is known to respond to CO2 levels during leaf development. Current interest in the increasing concentration of atmospheric CO2 has stimulated much experimentation on the responses of plants to relatively short-term exposure in artificially high CO2 levels. Attempts to extrapolate from short-term to long-term responses raise fundamental questions concerning evolutionary change in response to rising global CO2 levels.
Source: http://www.sciencedirect.com/science/article/pii/0024406693900059

The Impact of Atmospheric CO2 and Temperature Changes on Stomatal Density: Observation from Quercus robur Lammas Leaves

Abstract – Published: 01 March 1993

A comparative study of leaves formed on shoots during the spring and summer (lammas) of Quercus robur from three contrasting geographical locations (Cardiff, Durham and London) gives a measure of the effect of temperature on stomatal density. This is of value in attempting to distinguish the effects of CO2 and temperature on observed stomatal density changes under different CO2 and temperature conditions through the Quaternary. These leaves of normal and lammas shoots will have developed under similar CO2 levels but different environmental temperatures. Our results demonstrate that leaves formed under the warmer summer temperatures had reduced stomatal densities and indices from all sites, compared with their spring counterparts. This trend was also detected from measurements of spring and summer leaves made upon herbarium material collected from the same tree in 1840.
Source: https://academic.oup.com/aob/article-abstract/71/3/231/203377

And Ferdinand, cease with your “badmouthing” of Dr. Tim Ball, because he was 100% correct when he stated this, to wit:

What do the Ice Core Bubbles Really Tell Us?

The short answer to the question posed in the title to this article is virtually and practically nothing

Jim Masterson
Reply to  M Courtney
January 21, 2018 10:54 am

Ferdinand Engelbeen
January 20, 2018 at 12:09 pm

Thus if you see that stomata data in average differ from the average in ice cores over periods longer than its resolution, the stomata data need to be recalibrated towards the ice cores, not the opposite…

That would IMO constitute data tampering. Let each (ice cores and stomata data) stand on their own grounds and procedures.


Reply to  M Courtney
January 21, 2018 11:18 am


My information on stomata data was directly from a specialist (Van Hoof) on that matter, but if that isn’t up to date anymore, so be it. Anyway, as your sources show, stomata data are influenced by temperature during growing season, thus not only a CO2 proxy…

If you accept that CO2 can migrate from lower levels to higher levels without the help of pressure differences, then you may be right that CO2 data from ice cores are unreliable. My opinion is that it is physically impossible…

Reply to  M Courtney
January 21, 2018 11:44 am

Jim Masterson,

Ice core CO2 and other data are direct measurements, be it smoothed over some (long) period.
Stomata data are CO2 proxies which need to be calibrated against direct measurements. In general, ice cores, firn and Mauna Loa CO2 measurements are used to calibrate the stomata data over the past century.

Stomata data have a better resolution than most ice cores and show more variability. That may be real or may be from the fact that stomata data reflect local/regional CO2 levels over land, not “background” CO2 as is measured in 95% of the bulk atmosphere.
Ice cores from Antarctica show averaged global background CO2 levels over the periods of their own resolution. If the average CO2 level from stomata data over the same period differs from the average measured in ice cores, the problem is in the stomata data, as these are heavily influenced by local CO2 levels which may have changed over time, independent of global CO2 levels.
Take the situation at St. Odiliënberg, south Netherlands over the past 1,000 years, one of the main sampling sites for stomata data: in the main wind directions: marshes changed into forests, open sea was “poldered” to arable land, increasing population, increasing agriculture, late centuries industrialisation and increasing traffic…
In addition, nobody knows how the main wind direction changed in the colder Little Ice Age (at least in winter more East than the current “normal” SW)…

Samuel C Cogar
Reply to  M Courtney
January 23, 2018 4:43 am

Ferdinand Engelbeen – January 21, 2018 at 11:18 am


If you accept that CO2 can migrate from lower levels to higher levels without the help of pressure differences, then you may be right that CO2 data from ice cores are unreliable. My opinion is that it is physically impossible…


And just what was you deviously concocted reason for posting the above asinine conjecture. Do you really believe that your posting of the above comment will influence the world’s populace into thinking/believing that I am nothing more than a science “dumbass” of highly questionable character?

Ferdinand, please tell me, exactly what is the pressure differences between the 1,003 foot depth in the glacial ice ……. and the 1,006 foot depth in the glacial ice?

Please tell me, Ferdi, how is it possible for that extremely slight pressure difference in that 3 feet of glacial ice to be capable of forcing those CO2 molecules vertically through solid ice.

CO2 molecules are also subject to the “force of gravity”, ya know.

Ferdi, the BIG question is, which no one has addresses, …….. what were the local climatic conditions when the CO2 was being entrapped in snowfall on the surface?

Reply to  M Courtney
January 23, 2018 10:11 am


If you believe that Dr. Ball is right, then you also believe that Dr. Jaworowski was right by implying that CO2 can migrate from low to high levels.
As you clearly known that this is physically impossible, then why do you think that Dr. Ball is right by accepting what Dr. Jaworowski says?

Samuel C Cogar
Reply to  M Courtney
January 23, 2018 11:55 am

If you believe that Dr. Ball is right, then you also believe that Dr. Jaworowski was right by implying that CO2 can migrate from low to high levels.

Ferdinand, I have now read Dr. Ball’s above commentary twice, two times. And nowhere within that commentary could I find where Dr. Ball had even made mention that Dr. Jaworowski had claimed that “CO2 can migrate from low to high levels in glacial ice”, ….. and if Dr. Ball made no mention of said in his commentary …… then how in hell could he possibly agree.

Ferdie, it is devious, dishonest and disingenuous of you to post unsupported claims about what I, Sam C, believe or don’t believe …… based solely on what you think a 3rd party believes.

How many other people have you similarly “trampled on” to achieve your current career status?

Reply to  Ferdinand Engelbeen
January 20, 2018 11:57 am

Process Engineer in the science of Ice core chemistry?

Reply to  Sparky
January 20, 2018 12:46 pm

Sparky, if you go read FE’s links like I just did you will discover he has studied this stuff in considerable depth, including a long discussion of the problems with Jaroworski’s critiques.

Reply to  Sparky
January 20, 2018 12:46 pm


Neither was Dr. Jaworowski for CO2 in ice cores.

Before I comment on any subject, I try to read all sides of the discussion. I have read the objections of Dr. Jaworowski and I have read the work of Etheridge e.a. at three different ice cores at Law Dome, which imply a rebuttal to all of Jaworowski’s objections.
Etheridge used three different drilling techniques, wet and dry. measured CO2 top down from ambient, layer by layer in firn down to bubble closing depth. At that depth he measured CO2 in air from still open bubbles and from ice out of already close bubbles with the same equipment (GC), no differences found.

On the other side, Dr. Jaworowski made such impossible errors, that I can’t believe anything from him without thoroughly checking first. Like as above already said: CO2 migrating from inside to higher levels outside the ice, while O2/N2 don’t do that (at least not that fast: CO2 levels are ratio’s in air, not absolute values).

Another one: he accused Neftel of “arbitrarely” moving CO2 measurements some 83 years to align the CO2 levels from the Siple Dome ice core with the data of Mauna Loa. He simply used the wrong column in Neftel’s data table: the age of the ice instead of the average gas age. When I mailed him about that point, he answered that there is no difference between ice age and gas age, as most ice cores have multiple melt layers, which isolate the air from any further exchange with the atmosphere. Neftel clearly stated in his report that there was 1 (one) melt layer at some depth and that was accounted for in estimating the average gas age…

And so on. Sorry, Dr. Jaworowski was not a very reliable source of information about CO2 in ice cores…

Mike McMillan
Reply to  Sparky
January 20, 2018 10:21 pm

FE: … such impossible errors, … Like as above already said: CO2 migrating from inside to higher levels outside the ice, while O2/N2 don’t do that …

Below a certain level, the O2 and N2 start forming clathrates, so they’re locked in, unable to migrate, while the CO2 is still free to diffuse into and through the ice. The oxygen forms clathrates more readily than nitrogen, so the N2/O2 ratio in the bubbles climbs steadily as you descend, until you reach the bubble free region. I don’t remember the source paper for this chart:


and a chart for clathrate formation from Ikeda et al. 2000:


Reply to  Sparky
January 21, 2018 2:16 am


CO2 (and CH4) clathrates are formed before O2 and N2 clathrates, thus any migration of CO2 is stopped before O2/N2 clathrates are formed. There are even thought experiments to pump CO2 to the sea floor, where it may replace CH4 in sea floor clathrates…

Further, ice cores are put in storage at -20ºC for at least a year after drilling for relaxation: the ice expands again some 50%, decomposing a lot of clathrates. At measurement time, the ice is grated under vacuum and most clathrates left decompose, even violently.

Alternatively all ice is sublimated, leaving no clathrates at all and cryogenically separated into the different components, measured by mass spectrometry. The grating technique gives the same CO2 levels as the sublimation technique, except for single small samples at the clathrate transition zone. If one averages the measurements over a core length of 10 cm, the CO2 levels are “normal” again:

Samuel C Cogar
Reply to  Sparky
January 23, 2018 5:02 am

Ferdinand Engelbeen – January 21, 2018 at 2:16 am

Further, ice cores are put in storage at -20ºC for at least a year after drilling for relaxation: the ice expands again some 50%, decomposing

Ferdi, and during that horrendous pressure change and expansion of the ice cores, how far and which way did the CO2 molecules that were entrapped in the ice cores decide they wanted to migrate, ….. up the core, down the core or out of the core?

Reply to  Sparky
January 23, 2018 11:26 am


During relaxation, many air bubbles are reformed at the same place as they originally were. No measurable migration through the ice in periods of decades. If the expantion leads to cracks (as Dr. Jaworowski alleges), the cracks may reach the surface of the core and air can escape to the outside world as the inside pressure at that moment probably is higher than the outside. The escape may be in the same ratio for CO2/air as in the original bubble, or N2/O2 first as that are clathrates which form and decompose at higher pressures than for CO2. In that case, one will find higher CO2 levels in the remaining bubbles. Or if the ratio remained the same, once the inside and outside pressures are the same, the higher CO2 levels in the outside world can start to increase the inside ratio’s.

Thus whatever Dr. Jaworowski claims, every single problem in ice cores leads to too high CO2 levels in the measurements, not too low

Samuel C Cogar
Reply to  Sparky
January 23, 2018 12:57 pm

Ferdie, your above comment appears to me to be little more than “weazelworded” rhetoric that is intended to bedazzle the science illiterates who happen to read it.

Extrapolating 2 or 3 CO2 molecules detected in glacial ice……. to 10’s or 100’s of CO2 molecules ESTIMATED to reside in ancient near-surface atmospheres …… is both asinine and stupid ….. and also idiotic.

Ferdie, after spending tens-of-billion$ of dollars on ice core CO2 research in Antarctica, ….. just what the hell have they accomplished that is beneficial for the betterment of mankind?

James R McCown
Reply to  Ferdinand Engelbeen
January 20, 2018 12:40 pm

Ferdinand, even you know that air samples from ice cores from the Law dome in Antarctica have a resolution of at least several years. And yet some researchers have used Ethridge’s data as if it were discrete annual data of CO2 (or CH4 or N2O) atmospheric concentrations. I have posted on this previously here on WUWT. The following papers (and several others) have been published in climate journals claiming to have found a statistical correlation between greenhouse gas concentrations and annual temperature data, using that ridiculous ice core data.

Kaufmann, R. K., Kauppi, H., Mann, M. L., and Stock, J. H.: Does temperature contain a stochastic trend: linking statistical results to physical mechanisms, Climatic Change, 118, 729–743, 2013.

Kaufmann, R., Kauppi, H., and Stock, J. H.: Emissions, concentrations and temperature: a time series analysis, Climatic Change, 77, 248–278, 2006a.

Kaufmann, R., Kauppi, H., and Stock J. H.: The relationship between radiative forcing and temperature: what do statistical analyses of the instrumental temperature record measure?, Climatic Change, 77, 279–289, 2006b.

Kaufmann, R., H. Kauppi, and J. H. Stock, (2010) Does temperature contain a
stochastic trend? Evaluating Conflicting Statistical Results, Climatic Change, 101, 395-405.

Reply to  James R McCown
January 20, 2018 12:56 pm


The best ice core data we have are from three Law Dome cores with a resolution of less than a decade. Not yearly, but fine enough to show the huge trend in CO2 (and other GHGs, also CFC’s and other modern gases).
I doubt that one can find a srong correlation (there is always some with two upgoing trends…), as CO2 goes steady up since ~1850, while temperature goes up, down, up, flat,… in periods of about 30-40 years.

Anyway, the past temperature/CO2 changes in long term ice cores shows little influence of CO2 on temperature, temperature seems the driver for CO2 levels, hardly reverse…

Reply to  Ferdinand Engelbeen
January 20, 2018 7:15 pm

Ferdinand, I have been wondering about the use of carbon isotope ratios within samples from ice cores as an atmospheric temperature proxy. The stable isotopes show up in CO2 and the ratios are critical in interpreting C-14 dating results. Since C4 and CAM cycle plants metabolize CO2 with different carbon isotopes some differently from the way C3 plants do, the shift in the stable isotope ratio is used to adjust C-14 dates based on the type of plant providing the carbon. Would those isotopes go into solution in ocean water differentially based on atmospheric temperature?

Reply to  Duster
January 20, 2018 8:06 pm

In principle, any change in physical state (gas, liquid, solid) or chemical reaction will produce some isotopic shift in the species involved, and that shift will carry some degree of temperature sensitivity. Nobel prize winner Harold Urey did the basic mathematics in the late 1940s.

Reply to  Duster
January 21, 2018 2:28 am


Carbon isotopes indeed show a shift when transferred from water to air and back. Both ways show less of the heavier (13C) isotopes after transition, depending of temperature. If that is enough to be used as temperature proxy is questionable, as the 13C/12C ratio is influenced by lots of chemical, physical and biological reactions on this earth, thus hard to separate the origin of the changes. Add to that the huge influence of human emissions with low-13C carbon (-1.6 per mil since 1850) and that is a lost case.

The latest interesting finding is the use of noble gases to detect the full ocean temperature from the changed ratio’s in ice cores, as the solubility curve of the different noble gases is different with temperature. It was discussed here some time ago…

Reply to  Ferdinand Engelbeen
January 21, 2018 11:32 am


Of all the good contributors here, I think you are no less than the most senior expert of all concerning the carbon cycle.

It would be very interesting if you could write your own complete articles on this subject, which Anthony would hopefully accept, rather than just debunk the postings from others.

After all, the most important question in the whole climate debate is probably how much CO2 we can continue to emit if we want to stabilize the atmospheric CO2 level. I would appreciate to see a thorough analysis on this topic.


Reply to  Jan Kjetil Andersen
January 21, 2018 12:07 pm


Was already done some years ago with hundreds of comments and extended later with more information… Maybe a new update is necessary, as this kind of discussions pops up every now and then… Besides my comment on the objections by the late Dr. Jaworowski, I haven’t made (yet) a specific page on ice cores. With this discussion, we may have enough stuff to make one…

Already from 2010:
and from 2015, the theory behind the increase:

A comprehensive overview of the same is on my family pages:
and about the theory:

Reply to  Jan Kjetil Andersen
January 21, 2018 12:21 pm

Jan, forgot to add:

If you want to stabilize CO2 levels at the current mark, all you need to do is halve the current emissions and it is fixed. That would be economical suicide, but who cares…

The 60 years of CO2 data show a quite linear sink rate for CO2 in ratio to the extra CO2 in the atmosphere above (dynamic) equilibrium (“steady state”) between the ocean surface and the atmosphere for the current average ocean surface temperature (about 290 ppmv). From the about 9 GtC human emissions, half of it (as mass, not the original molecules) is absorbed by the (deep) oceans and vegetation. There is no sign of any slowdown, to the contrary, despite what the IPCC says. The IPCC expects a slowdown as the (deep) oceans may saturate somewhere in the far future. That is based on the Bern model, which still need to be validated, but is based on 10-20 times the current emissions…

Thus if you halve the current CO2 emissions, the net sink rate will remain the same for the same 110 ppmv extra in the atmosphere and no further increase will be the result.

Reply to  Jan Kjetil Andersen
January 21, 2018 12:48 pm

Thank you for the links to the articles.

It is interesting to see that you think that a halving of current emissions is enough to stabilize the CO2 level.

I agree that it make sense since, about half of the current manmade CO2 emissions stay in the atmosphere and the other half goes to the ocean and biosphere. If we remove one half the level should stabilize, at least in the short run.

However, have you considered whether this is enough in the long run? Could the biosphere and the ocean surface water be saturated with CO2 after some years, so we then would have to reduce the emissions further?


Reply to  Jan Kjetil Andersen
January 21, 2018 7:01 pm

It’s nonsense. The ice cores cannot be validated, and temperatures drive the rate of change of CO2.


Reply to  Jan Kjetil Andersen
January 21, 2018 11:23 pm

Thre is a positive feedback loop between CO2 and temperature. If one of them for whatever reason goes up, that drives the other to also go up.

Reply to  Jan Kjetil Andersen
January 22, 2018 4:24 am


That the sinks saturate is the main fear of the IPCC: for each compartiment they have a saturation limit, which makes that a lot of CO2 remains in the atmosphere for very long periods up to indefinitely. That is based on the Bern model:
The Bern model itself is based on burning enormous amounts of fossil fuels: 3,000 to 5,000 GtC, while all human emissions up to now are about 370 GtC…

Based on the observed sink rate, there is no sign of a reduced sink rate: the same ratio over the past 60 years. The Bern model is right for the ocean surface, which saturates at about 10% of the change in the atmosphere, but wrong for the deep oceans, where saturation is not in question for the long foreseeable future. Not at all applicable for the biosphere, where more CO2 only is good for more growth…

Reply to  Jan Kjetil Andersen
January 22, 2018 5:43 am

Jan Kjetil Andersen @ January 21, 2018 at 11:23 pm

“Thre is a positive feedback loop between CO2 and temperature.”

Not possible. Due to the integral relationship between temperature and CO2, that loop could not be stabilized even by T^4 radiation. The only possibility is that the aggregate temperature response to CO2 is negligible or even negative. This does not mean that CO2 has no warming potential – it clearly does. But, it is immediately offset by other dynamics which have not been given adequate consideration.

Reply to  Jan Kjetil Andersen
January 22, 2018 6:33 am


I have had a lot of discussions with Bart about his theory: he thinks that the increase of CO2 is the result of the integration of natural emissions in ratio with a temperature offset of a baseline. Mathematically possible, as one can match any two slopes if not too far from each other if you take the right offset and factor.
What is sure is that the variability of the temperature rate of change is the cause of the variability of the CO2 rate of change. Where Bart is wrong is that the slope of the temperature causes the slope in CO2 rate of change. That relationship is entirely spurious, as the variability is caused by vegetation, but the slope is not by vegetation: that is a proven sink for CO2…

Problem is that it does violate every single observation over the last 60 years, while human emissions fit them all.
As most was said (too) many times, I don’t want to reopen that discussion here again. If you have a lot of spare time, have a look at the discussion here:

Reply to  Jan Kjetil Andersen
January 22, 2018 6:51 pm

It matches both the trend and the variation in the rate of change for the past 60 years. It’s a slam dunk.

Tom Bjorklund
Reply to  Jan Kjetil Andersen
January 22, 2018 7:20 pm

Bartemis, your correlational relationship does not show any type of causation. Please get back to us when you can show causation.

Tom Bjorklund
Reply to  Jan Kjetil Andersen
January 22, 2018 7:30 pm

Bartemis says: “temperatures drive the rate of change of CO2.”

However the graphic he shows us does not disprove the alternative claim that the rate of change of CO2 drives temperature.

In simplistic terms, the relationship that is shown by the graph does not identify which variable is dependent and which is independent.

Reply to  Jan Kjetil Andersen
January 23, 2018 2:28 am

Tom Bjorklund,

You have nailed the problem with Bart’s theory in our many years discussions with him:

Bart compares the slope and variability of temperature and the derivative of the CO2 increase. That has zero lead or lag, as T and dCO2/dt are fully synchronised.

How is that possible? Any sinusoid shows exactly the same waveform when the derivative is taken, only shifted back some 90º. Around the slope of CO2, the small (+/- 1.5 ppmv) changes lag T changes with about 90º. The backshift by taking the derivative of CO2 synchronises both T and dCO2/dt.

The real lead-lag is between T and CO2 (for the variability around the trend, not for the trend itself!) and as derivatives between dT/dt and dCO2/dt. The problem for Bart’s theory is that dT/dt has exactly the same variability as T, but zero slope, thus is not the cause of the slope of dCO2/dt, but the cause of most of the variability. The slope of dCO2/dt is almost completely caused by the slightly quadratic increase of human emissions, leading to a slightly quadratic increase of CO2 in the atmosphere and a linear slope in the CO2 derivative:


The human emissions curve is about twice the CO2 derivative curve but near linear without much variability.
The δ13C derivative curve is near exactly opposite to the CO2 derivative, which proves that (tropical) vegetation is the dominant reactant to temperature variability.

Reply to  Jan Kjetil Andersen
January 23, 2018 6:22 pm

Tom Bjorklund @ January 22, 2018 at 7:30 pm
“However the graphic he shows us does not disprove the alternative claim that the rate of change of CO2 drives temperature.”
Incorrect, as it is a rate of change relationship. The rate of change of CO2 cannot be driving temperature. If it were, then CO2 could rise to any level, but temperatures would revert to their original level once the rise stopped, and that would be absurd.
No, the relationship firmly establishes temperature as the cause, and CO2 as the effect.

Reply to  Jan Kjetil Andersen
January 24, 2018 6:42 am


If it were, then CO2 could rise to any level, but temperatures would revert to their original level once the rise stopped, and that would be absurd.

Interesting, as your theory shows the opposite absurd rule: any small offset of temperature from an arbitrary baseline will increase CO2 indefinitely and a return to the baseline only stops the increase, leaving the extra CO2 in the atmosphere indefinitely

You can’t deduce the above from the T-dCO2/dt relationship as both are fully synchronised. A sudden shot of CO2 has an immediate effect on temperature. That effect remains and doesn’t reverse the temperature, only stops the increase once the CO2 increase stops. If we may use some fantasy…

In reality CO2 variability follows T variability and dCO2/dt follows dT/dt variability, both with a lag. That makes that T has zero influence on the slope of dCO2/dt, only a small influence on CO2 levels.

Reply to  Jan Kjetil Andersen
January 24, 2018 11:44 pm

There is nothing absurd about it at all. It is a very commonplace dynamic over finite time intervals. I’ve provided a model to you previously that shows how it can come about.

Reply to  Jan Kjetil Andersen
January 25, 2018 12:42 am


Here we go again…

No matter how nice your theory, it is quite simple: if a theory violates even only one observation, the theory is wrong.
Your theory violates every single known observation and your only defence is that all these observations must be wrong.

End of discussion.

Reply to  Jan Kjetil Andersen
January 26, 2018 10:03 pm

It violates no observation. Your narrative is not an observation.

Reply to  Ferdinand Engelbeen
January 21, 2018 12:16 pm

Mr. Englebean
Your comments add a lot of value to this blog.
I think more people should take the time
to tell you that.

Reply to  Richard Greene
January 21, 2018 12:17 pm

I meant website, not “blog”

Reply to  Richard Greene
January 21, 2018 12:25 pm


Thanks a lot, I do appreciate the appreciation…

Reply to  Richard Greene
January 23, 2018 8:46 am

Mr. Englebeen
Sorry I spelled your name wrong.

A few days later, I want to add,
and hopefully not insult Dr. Ball
whose article instigated your comments,
that the sum of all your comments
adds up to a very good article, actually
better than the original article!

However, concerning ice cores,
i wonder if they were worth the expense
of digging up, and analyzing.

I think we already knew the CO2 level
has gone up a lot from man made emissions,
although I think that is good news,
while a few billion people disagree!

And every proxy, perhaps excepting the tree rings
or frogs, or whatever it was that Dr. Mann used,
tells us the temperature
is always changing
(at least at one point
on our planet
where the proxy was located

If we already knew Co2 was rising, and
the average temperature is always
changing, then what new knowledge
was gained from the ice cores?

My only answer, maybe, is that temperature
peaks led CO2 peaks by several hundred years,
and I’m not sure the global warmunists
will even acknowledge that today.

We did not learn what the future temperature will be.
(no one knows)

We did not learn whether CO2 became the
climate ‘controller’ after 1940 (it didn’t).

And we certainly did not learn what a “normal” average
temperature is, or what a “normal” CO2 level is,
assuming “normal” can be defined on a planet
not in thermodynamic equilibrium.
(it can’t)

So exactly what was learned from ice cores,
that was not known, or at least strongly suspected,
before the drilling ?

My primary concern is always about the money
spent, and what was gained from the spending,
versus alternative scientific uses for the money … or not
spending so much of the taxpayers’ money on
what passes for climate science these days.

For all the billions (trillions) spent
on climate change, and alternative energy
subsidies, I still have no idea what the future
climate will be, and I find absolutely no real science
at all behind the dubious claim that
doubling the CO2 level in the atmosphere
will cause +3 degrees C. warming,
much less runaway warming.

I have a BS degree, but see no value from it
in the fairy tale world of modern climate “science”,
where computer game predictions of the future
climate, which used to be called wild guesses,
now pass for real science.

My climate change blog is blow.
Liberals must stay away,
or elks.


Reply to  Richard Greene
January 23, 2018 11:52 am


My (Flemish/Dutch) name’s spelling seems quite difficult for Anglo-Saxons (just like the impossible non-phonetic English spellings for us…). Even if we have some common roots (Saxons and Franks were the main ingredients here with some flavour of the Norse at the coasts. No problem with that, have seen worse…

Where I disagree is that ice cores are not worthwile digging up.

First to satisfy human curiosity, the same reason why we explored the depths of the oceans, the tops of the Himalaya, sent a man on the moon, and an explorer to Mars.
Second, because climate of the past had a few, sometimes nasty, surprises like the sudden cooling in less than a decade of the Younger Dryas. And still a lot of questions left like as why the long time that CO2 remains high in the atmosphere while the temperature already dropped to a minimum at the end of the interglacials…

You have a nice collection of old weather/climate disasters!

Reply to  Ferdinand Engelbeen
January 21, 2018 1:59 pm

So what are you neglecting here, Mr. Engelbeen?
“Dr. Jaworowski made several statements which were physically impossible, like CO2 escaping (preferentially?) via cracks in the ice to the outside world, while inside one found 180-280 ppmv and the outside air in the labs was at least 380 ppmv (or more in enclosed rooms with people in it).”

It is interesting to see that suddenly the CO2 we breathe out is taken into consideration. Nowhere else in the thousands of pages of IPCC papers I’ve read this is mentioned. It’s all about the evil CO2 from fossil sources.

You are also neglecting the fact that air from the lab will enter the ice and vice versa. The same processes will take place while the cores are transported from Antarctica and to the lab. Some of the early Antarctic cores were transported across equator, and most of the ice in the cores melted. There was also a problem with contamination from drilling fluid, which resulted in the discarding of high CO2-values, which led to the ice cores mostly missing the high atmospheric values around 1940 (380 ppm in 1942, see Beck 2007 in EE).

I have noticed that Segalstad who did research with Jaworowsky is still of the opinion that CO2 locked into ice bubbles does NOT remain unchanged as time is passing, since there is 20 physical-chemical processes taking place in the ice.
Z. Jaworowski, T. V. Segalstad og N. Ono 1992, Do glaciers tell a true atmospheric CO2 story? Science of the Total Environment, 114, 227–284, http://www.CO2web.info/stoten92.pdf

With so many errors in your many comments here, why should we have any confidence in the general quality of your comments?

As for the Petterkoffer method (+/- 3%, not +/- 10 ppm), from 1935 an improved Pettenkofer method had a precision of 0,33 %.

Reply to  jarlgeir
January 22, 2018 6:18 am


The CO2 we breath out is only important in closed rooms, as high levels means less oxygen left…

It may have an influence on ice cores, if there was any exchange between the air bubbles inside and the air outside. In all cases the levels inside would increase, as we measure 180-300 ppmv or gas ages up to 1850, while the outside world now is 410 ppmv. If there was any migration, levels would increase over time during storage, but that was never observed. That is the whole point where I differ in opinion with the late Dr. Jaworowski and Segalstad: any contamination from the outside world would increase the measured CO2 levels, not decrease.
The same for contamination with drilling fluid: Neftel said that high CO2 levels were found with cracks and drilling fluid in the Siple ice core. When ice was taken from the same parts, avoiding cracks, the CO2 levels were in line with the other measurements. Dr. Jaworowski simply ignored that warning and accused Neftel of removing these high values, as if these have any value. In later years a new core was drilled at Siple Dome which didn’t show any extreme high values.
Are you of the opinion that one should retain any outliers by contamination of which the origin is known? I don’t think so…

Globally, the CO2 we breath out has no effect, as all what we use as carbon food was months to years before taken out of the same atmosphere where we move it in again. The overall effect of the biosphere is more sink than source: plants take more CO2 out of the atmosphere than they release by decay and its use as feed and food. That is why CO2 from breathing is neglected: it is only part of the natural carbon cycle.

There were no high CO2 values around 1942. The high values found by the late Ernst Beck were the result of CO2 measurements over land, contaminated by local sources. These high values are not found in several ice cores, stomata data, coralline sponges of any other available proxy.

The average accuracy of the old methods indeed was +/- 3%, or +/- 9 ppmv on a level of 300 ppmv, more accurate methods may have existed, but I never have seen any calibration results in the field (with one exception, which was a disaster…), which is not only a matter of method and equipment, but also a matter of fresh reagentia and skill.
Even if the method was better than 1 ppmv with fresh reagens and skilfull people, that doesn’t help you if one measures CO2 three times a day, at the edge of a forest where the day level is 250 ppmv and the night level 500 ppmv. That was the case for the longest series of Beck’s compilation which largely caused his non-existing 1942 “peak”.

I am not aware of any errors I have made here, and if so I have no problem to admit that I am wrong. Thus please cite my exact words and why I am wrong. That doesn’t include citing the work of Jaworowski and Segalstad which was already refuted in… 1996 by Etheridge e.a..

Reply to  jarlgeir
January 23, 2018 9:03 am

The Pettenkoffer method provided excellent measurement data, but not for locations that would represent a global average.

However if it was possible to replicate the same or similar locations of some of the old measurements, using similar measuring instruments, it might provide useful information on the Co2 change (and whether ice core proxies from one place, are really useful to estimate the global average CO2 — it might seem like the ice cores were good proxies for that, but that’s just an assumption, not a proven conclusion).

+/- 10 ppm would be enough accuracy

There are plenty of underseas volcanoes emitting Co2, but no one cares about them
except for the one(s) near the Antarctica peninsula that causes about 2% of Antarctica to be warming
while the other 98% is not. That 2% gets all the attention — just the warming itself — almost no one mentions the underseas volcano.

It is good that Mr. Jaworowski challenged the accuracy of the ice core data — good scientists should always challenge the “consensus”, but it appears he was wrong, and i am convinced Mr. Englebeen here has proven that.

However, we don’t need ice cores to tell us we have, and are, adding CO2 to the air.

What we need is common sense to see that adding CO2 is good news.

And if there is any warming at night in high latitudes, from the added Co2,
then that is good news too.

I believe a debate with Mr. Englebeen on ice cores
is likely to be a sure loss, and would move on.

the coming climate change catastrophe
is a leftist fairy tale
that has nothing to do with ice cores.

Reply to  jarlgeir
January 23, 2018 12:14 pm

Richard Greene,

Indeed I would like to see a real life comparison of the old wet methods with the modern instrumentation at the same places as measured today.

There were two historical places of interest: a base at Antarctica (the current Mac Murdoch?). Some measurements were quite high, but as the oxygen measurements at the same time were several % too low, that points to contamination. Further, the equipment was broken in a storm during the return trip, so couldn’t be recalibrated on return.
The other was at Barrow, currently one of the NOAA base stations for measuring background CO2. That was used to test CO2 in exhaled air of the researchers there. As that is 20-40,000 ppmv, the equipment was calibrated with outside air. If the reading was between 200-500 ppmv, the equipment was deemed fit for purpose… With a accuracy of +/-150 ppmv, not much can be concluded…

What is a good indication is that all measurements done on board of seaships or coastal with wind from the seaside were all around the ice core CO2 levels of the same period.

January 20, 2018 11:33 am

It’s too bad that trees don’t grow on glaciers. Then they could use tree rings to cross-check the air bubbles in the ice for a really stupendously accurate temperature record. Nobody could argue with it then. 🙂

Curious George
Reply to  daveandrews723
January 20, 2018 11:38 am

Hmmm .. could a permafrost be used for this purpose?

Reply to  Curious George
January 20, 2018 11:56 am

No, permafrost doesn’t accumulate in annual layers like a glacier and the active layer melts every year. However in Antarctica there has been som work done on very old ice buried in permafrost that in some cases maybe several million years old:


Reply to  daveandrews723
January 20, 2018 8:26 pm

Trees can be found under the glaciers when they melt, the modern taxa thereof demonstrating clearly that periods of warmer temperatures than today have taken place since the LGM.

Reply to  daveandrews723
January 23, 2018 9:20 am

Maybe a penguin count would be useful too?

How about you take charge of that,
Mr. Andrews?

dave andrews — you use your real name
on the internet? and tell good jokes too.

That puts you in the top 1%.

People think Richard Greene is my real name,
but it’s actually Melvin Q. Schitzelwitz,
and don’t ask what the “Q” stands for.

And my jokes are so bad,
I have to tell people a joke is coming,
type the joke, and then tell them
I just told a joke …
and they still don’t get it.

WUWT is the best climate website in the world.
Anthony Watts is a climate skeptic legend,
for this website, and his climate station study.

The only problem here is the Moderators,
(actually one person who pretends to be a “team”)
who is obviously a robot,
with fantasies of being
the next Hemingway,
although the only resemblance
concerns drinking a lot of alcohol
and spending days and nights
sitting on bar stools …
and the result is a person
who would not
recognize a good joke
if it walked over to them
and sat in their lap.

This post is moderator bait.

One must keep the Moderator well fed,
or he (she? it?) will get bored
and cranky. And there’s nothing worse,
than a cranky Moderator.

Estimated time for deletion
of this post: Five minutes
after the Moderator sobers up.
heh heh
Love a good joke and a good drink, but I am not the moderator you are referring to! Mod

(Looked up the history to see that it is not spam or from the Onion website) MOD

January 20, 2018 11:50 am

Dr. Ball ==> I am not a fan generally…..usually do not comment on your essays here out of respect for how you have been treated by the greater CliSci community and the press. Simply put, sometimes your essays are over the top, seemingly fueled by your justifiable anger, and contain the same types of faults contained in the unfair attacks on you by others.
This essay, however, is quite welcome, well done and informative and is nice contribution to the overall content here at WUWT. Thank you.

Reply to  Kip Hansen
January 20, 2018 12:38 pm

Kip you wrote:
“Simply put, sometimes your essays are over the top, seemingly fueled by your justifiable anger, and contain the same types of faults contained in the unfair attacks on you by others.”

Can you cite examples Kip? I cannot, and I read everything posted on wattsup by Dr. Tim Ball.

Given the many death threats leveled against Tim by imbecilic global warming alarmists, I could understand if he was somewhat testy – but I have not seen that conduct in his essays and responses. Others of us are more inclined to lash out from time to time, especially if we believe the opposing party is deliberately acting in a deceitful manner (which is not uncommon).

I did receive one rather crude threat in 2002, and I can assure you that it was disturbing at the time. Now I view it as ridiculous, and feel that I am under-appreciated by the thugs of the warmist camp (“Don’t I deserve more threats? What am I, chopped liver?”). 🙂

Regards, Allan

January 20, 2018 7:04 pm

Allan ==> I’m not sure why you are replying to my comment address to Dr. Ball. “Can you cite examples” …Dr. Ball knows, and you could find out my using the search tool here to read all his previous essays here.

January 21, 2018 8:41 am

Hello KIp.

I consider Dr. Tim Ball a friend, having met him circa 2001 and having corresponded with him many times. I do feel somewhat defensive towards my friends when they are unjustly attacked, and that is why I responded to you.

To be clear, you have not provided any examples, so I am still skeptical of your allegations. To my knowledge, Tim has managed himself as a gentlemen despite many unwarranted personal attacks, so I reject your comments as unwarranted.

Unlike Tim, I do “lash out” from time to time, especially when I perceive that warmists are deliberately acting in bad faith, which is not uncommon.

The science of global warming is not settled, but from what we know today this is NO real catastrophic global warming caused by increasing atmospheric CO2.

Regards, Allan

Reply to  Kip Hansen
January 21, 2018 12:25 pm

Kip “I’m not a fan” Hanson:

If you want to attack Dr. Ball’s essays
on this website, then please provide
specific details of what is wrong,
rather than launching,
a generic character attack,
similar to what leftists do
to climate skeptics.

It’s puzzling for you to make
a generic character attack on Dr. Ball
and then complement his latest article.

Sort of like this
back-handed complement
for you:
Your 2017 essay on obesity
was trash,
for the many detailed,
long-winded reasons
I provided at the time …
but your last series of articles here,
on sea level rise was excellent
— I thought Part 3 was the best article
at this website in 2017 and
said so.

I think you should combine all four
sea level articles into one summary article,
and that could be the best article here
in 2018.

When you insult someone first,
the complement that comes later,
does not mean much.

Sort of like my wife
telling me the current
issue of my eight-page
economics newsletter
was unusually good,
because she usually
falls fast asleep
while reading my newsletters !

What you see as “anger” in Dr. Ball,
I see as passion.

J Mac
January 20, 2018 11:56 am

Dr. Ball,
Thank You! There is much to digest in your post.

As a metallurgical engineer, trained in and dependent on measurements of elements within liquid and solid phases, I have always had a dubious sense of the legitimacy of glacial ice core gas chemistry determinations. Your post reinforces my skepticism and adds additional reasons for it ‘in spades’.

Side bar: There is something (can’t quite pin it down…) about the term ‘firnification’ that tickles my sense of humor! The wag that coined that term was a sly one.

January 20, 2018 12:02 pm

At last! – A good analysis of ice core results. My thanks.

For my part as a mere observer I was worried about these ice core results on two main accounts: First being the hockey stick presentations let loose on the media and secondly on the basic fact that CO2 is far more soluble in water than either of the other two gases. I was also convinced that these bubbles were not closed systems and were open to matters of diffusion physics (Fincks Laws?), high pressure and other factors of which I was not aware.
I did enquire about these matters; but was told that all was in hand.

It is indeed gratifying to me that we are now seeing this and now many examples of scientists challenging the 97% SatanicCO2 Meme which has so infected current science.

It will take time; but truth will out. Expect I will be gone by then. – Keep at it folks!!

January 20, 2018 12:09 pm

[snip – policy violation, something you have done multiple times here. Your comments are designed to drive traffic to your website, buy making an argument, taunting people here, and then providing multiple links to your theories. This is thread-bombing. We’ve deleted similar comments of yours in the past, but you don’t seem to be getting the message. Keep it up and you’ll find yourself permanently banned – Anthony]

Reply to  nickreality65
January 21, 2018 12:28 pm

nickreality65 has been a bad boy!

To pile on,
up your nose
with a rubber hose,
Mr. Nick

Moderator bait.

January 20, 2018 12:16 pm

In short, what happens is:

The gas in the ice layers already has a 500 year smoothing applied to it, and that smoothing displaces the gas data from the ice layer data by a few hundred years. (Known as the gas-age.)

And we know that the gasses in the bubble are not perfectly preserved, because dust contamination in the Greenland ice cores means that CO2 levels are not measureable. The CO2 HAS been altered while in situ.

However, there are other records that support the general structure of the ice core temperature record. These include:

Ice core methane, Chinese stalactites, Deep sea silt deposits, and Chinese dust doposits upon the Loess Plateau.

All of these tell the same story of large temperature/climatic fluctuations during the many ice ages, accompanied by equivalent CO2 fluctuations. But what these records cannot demonstrate, is that this correlation means causation. In fact we know that there is no causation by CO2 because during the ice ages:

When CO2 is high, the world cools.
When CO2 is low, the world warms.

If CO2 were the primary feedback agent, it is highly unlikely that this would happen. And so the entire AGW theory is based upon a false presumption. In reality, the primary feedback agent that controls ice age temperatures, is ice sheet albedo.


Reply to  ralfellis
January 20, 2018 4:12 pm

Let me try this again (Why do so many of my comments get sent to Mod Purgatory?? I don’t say anything outrageous or defamatory)

Thank you, ralfellis, for making this critical point:

“When CO2 is high, the world cools.
When CO2 is low, the world warms.”

It is bad enough for the AGW alarmists that temperature rises/falls BEFORE CO2 rises/falls. So time’s unidirectional arrow proves that T moves CO2, not the other way.

But worse, the measured records show repeatedly that COOLING begins when CO2 is at its peak value, and WARMING begins when CO2 is at its lowest value.

Quite simply, CO2 cannot have any significant effect on Temperature (It”s the other way around)

Why does this seem so difficult for so many to understand, even on the skeptic side?

Reply to  GeologyJim
January 20, 2018 4:43 pm

Because, GeoJim, at 3C ECS, CO2 only accounts for just a third of the warming from glacial to interglacial. (so there isn’t the claim that CO2 is the primary driver in the first place) That said, an ECS of 3C does seem a little too high. 1/3 is an awfully big slice of the pie when you consider all that tiny CO2 is competing with. The changes in water vapor alone are enough to dwarf the warming by CO2…

Reply to  GeologyJim
January 20, 2018 6:30 pm

Geology Jim Karl Popper very carefully described the boundaries of science and not science. As scientists since we can never do infinity tests on anything we can not declare a “Truth” simply that the theory has demonstrated skill and because of the repeated skillful results warrants confidence until a falsification is noted. At that time ” in science” the theory is abandoned or modified to improve skill. An example would be the earth centric solar system as contrary information was gathered the theory became increasingly complex until the moons of Jupiter could not be explained. However celestial navigation which requires the skillful prediction of the location of planets and stars works with the failed theory. Just a Newton’s theories work and are practical for most of us. Karl Popper was alleged to have removed any assertion that Freud and Maxist theories were scientific. And he placed these in the other outside of science. The unifying characteristic of ideologies is that unlike science an ideology requires each member to cede the determination of “TRUTHS” to the higher authority. They are limited to writings and discussions on the meaning of the “TRUTH* but not to question it. This would be heretical and in ideologies heritics are ostracised. With this background you have to realize that if a person on receipt of falsification modifies the theory to improve skill they remain within the parameters of science. If however they ignore the contrary information anf persist with a falsified theory then they exist outside science. This is a simple go no go gauge .if you are confused it is like arguing with a person whose “truth” as handed down to them is that the earth is 6900 years old. As scientists we have to manage the nature of the inquiry and the philosophical boundary. If the theory is falsified and isn’t changed the discussion is outside the boundary of science. You can step outside science to contemplate any ideology and personal belief systems and what politics should do based upon science but the ideologies have no place in science. This “crossing of the line” is the essence of the phenomina you are observing and it has to be called out.

Reply to  GeologyJim
January 21, 2018 12:34 pm

>>Because, GeoJim, at 3C ECS, CO2 only accounts for
>>just a third of the warming from glacial to interglacial.

But what they will not tell you, is that the albedo of dusty ice contributes orders of magnitute more warming than CO2 (when calculated locally). Easily 200 wm2, instead of the paltry 3 wm2 of CO2. And it so happens that each interglacial is preceded by a few thousand years of dust.


January 20, 2018 12:20 pm

Dr. Tim Ball, you are a good man, a very good one when at it…..
Never let any one telling you otherwise….

With all arguing and battling around in these threads and else where, you still shine through the goodness and strength of that all… human integrity….and steadfastness in duty and greatness…with no any doubt, if I may put it like that…A feat only matched by the good and the great…

It is so painful, if I have to put it this way…it is so painful that there is so few man like you Dr. Ball….

In my understanding the whole world will definitely be a better place if there for some luck it will happen to be more man like you Dr. Tim Ball…
But still we got to do and get through with what ever given…

Thanks a lot…


January 20, 2018 12:21 pm

I learn so much whenever I come to this website—and not just about climate science, but scientific principles and quotable quotes as well like the one right at the end from A. N. Whitehead. Skepticism is the hallmark of good science and I see that logic dictates interpretation (or at least, ought to.) To be labeled a climate skeptic is a badge of honor and pride for me. Thanks again for the excellent articles presented.

Intelligent Dasein
January 20, 2018 1:17 pm

A hockey stick is precisely what you would expect to see if you began with the assumption that CO2, which is quite soluble in cold water under high pressure, was diffusing through the glacier and being squeezed out over time.

Reply to  Intelligent Dasein
January 20, 2018 1:39 pm

CO2 is soluble under high pressure in water – yes, but I had to give up drinking Diet Coke because it is very bad stuff. I miss the fizziness. I guess I could get a spritzer, couldn’t ?

sy computing
Reply to  Sara
January 20, 2018 2:52 pm

“I miss the fizziness. I guess I could get a spritzer, couldn’t ?”

Try the SodaStream ma’am!


Reply to  Sara
January 20, 2018 6:45 pm

“thread-bombing” Learn something new everyday. Thanks Anthony.

Reply to  Sara
January 20, 2018 6:49 pm

Wrong place for above. It was going as a reply to a post Anthony snipped. I was going to second Soda Stream. I carbonate water and add fresh lemon or lime juice (no sweetener)

Reply to  Sara
January 20, 2018 7:34 pm

Perrier is good because it is naturally carbonated, just as champagne carbonates itself.

Tom Bjorklund
Reply to  Sara
January 20, 2018 7:58 pm

Perrier is not naturally carbonated.

” Both the water and carbonic gas are still harvested independently. They come from isolated points at different depths, within the same geologic formation. ”
Reference: https://www.nestle-watersna.com/asset-library/documents/p_eng.pdf

sy computing
Reply to  Sara
January 20, 2018 9:57 pm

“Perrier is not naturally carbonated.”

Really Tommy? You just must correct the lady? Is that it?

Make you feel better did it?

Mind yer biscuits mister or I may come over there and carbonate your Bjorklund to teach you some manners…

Reply to  Intelligent Dasein
January 20, 2018 1:46 pm

Intelligent Dasein,

If you have any knowledge of diffusion, you would know that it is impossible to “squeeze out” CO2 from 180-300 ppmv inside towards 400 ppmv outside nowadays. The opposite would be more likely. Moreover, where would the CO2 go under 2 km thick ice and with what speed at -40ºC, thus near zero water left to migrate in…

Reply to  Ferdinand Engelbeen
January 20, 2018 5:31 pm

If you have any knowledge of diffusion, you would know that it is impossible to “squeeze out” CO2 from 180-300 ppmv inside towards 400 ppmv outside nowadays.

That’s assuming that Henry’s Law holds for polar ice. There is some evidence that Henry’s law may not always apply.

For example, the assumption of Henry’s law for the CO2 solubility in ice may not be valid through all pressure ranges. page 692

The behaviour of CO2 in polar ice is indeed poorly understood.

Intelligent Dasein
Reply to  Ferdinand Engelbeen
January 20, 2018 6:49 pm

Diffusion isn’t the issue here. I am referring to mass transfer of a solute in briny layers of liquid water.

R Taylor
Reply to  Ferdinand Engelbeen
January 20, 2018 8:19 pm

Another reason to have confidence in the Antarctic CO2 record is the consistency of ranges and lags, relative to the deuterium and 18O records, through several 100,000 year cycles.

Crispin in Waterloo but really in Ulaanbaatar
Reply to  Ferdinand Engelbeen
January 20, 2018 11:37 pm


“Moreover, where would the CO2 go under 2 km thick ice and with what speed at -40ºC, thus near zero water left to migrate in…”

I think perhaps you missed the papers, at least one of which was dealt with here, describing the water that exists as a liquid within low temperature ice under pressure in Antarctica.

Ice at -40 is not entirely solid. There is liquid water within it. The mechanism I cannot fully describe so I won’t try. But it is not correct that there is zero water to migrate in. There are also little critters living in the water that permeates the ice. Nature is a bugger when it comes to explaining it in a simplified form, and never cooperates.

In view of this, several things you have posted above need revision. CO2 can contained and diffused though the ice, and the water within the ice, and also to form several gas and liquid products other than free CO2.

While I admire the efforts scientists have made to determine atmospheric gas composition using ice cores, I have much more faith in the stalagmites and clay deposition cores based on mineralogy, not trapped gases. I work with gases. They have a lot of troubles.

The major take-away from ice cores is that temperatures rise before CO2. After that, meh… Give me a nice, cool Chinese limestone cave populated by flappy-bats.

Reply to  Ferdinand Engelbeen
January 21, 2018 2:45 am


The solubility of CO2 in the water-like layer at the ice-air layer of the enclosed bubbles may not follow Henry’s law, but that is not important in an ice core of 3 km thick, as there is little difference in pressure between 2000 m ice and 2001 m ice. So any migration of CO2 at -40ºC would be through the crystal bounderies – as far a there is any liquid – and only driven by concentration differences.

It is simply impossible that low CO2 levels migrate to higher ones, without huge pressure differences. Even in the case that bubbles connect to the atmosphere through cracks while still under high pressure just after drilling, O2/N2 will escape first (and their clathrates will decompose first), as CO2 has a higher affinity to water/ice, effectively increasing the CO2/air ratio in the remaining air. Moreover, diffusion after long periods of storage would show increasing CO2 levels as the environment is lots higher in CO2 ratio. That is not the case at all.

If there are physical changes, one will always find higher CO2 levels, not lower as Dr. Jaworowski alleges.

Reply to  Ferdinand Engelbeen
January 21, 2018 2:59 am


The water-like layer is only at the ice-air border and is about 5 molecules thick at -40ºC. Not much CO2 to hide there and as the measurements are done under vacuum, that layer, including any CO2 is removed and water vapor is trapped at -70ºC, leaving even less CO2 at the surface.

The intercristalline border is disordered water, not liquid, and migration only may occur due to dust inclusions which are rare in inland ice cores of Antarctica. The same for bacterial inclusions, often in combination with dust, good for maximum 0.5 ppmv loss of CO2 in the Vostok ice core at the most contaminated depths.

Was a few years ago in China on a round trip and visited one of their caves. Indeed beautiful, but I have more doubts on the side influences (precipitation changes, carbonates dissolvation from different sources…) on stalagtite isotopes than in Antarctic ice cores…

Reply to  Ferdinand Engelbeen
January 21, 2018 6:16 pm

But aren’t we talking about CO2 trapped in ice cores that have been around for decades, centuries, millennia when hypothetically “outside” levels were far lower than they are today. Let’s face it, what is fact is that the ice and trapped gases, etc come from a dynamic real world environment. “They” have not been just sitting around all these years waiting for humans to use a coring drill. Like many things in the AGW world ice core data sounds like the gases have been “locked away” in some big freezer just waiting for us to defrost it.

Reply to  Ferdinand Engelbeen
January 21, 2018 6:28 pm

The bubbles seem like very good time capsules, perfect? No, but still very useful.
Sometime science is better than gut instinct.

Joel O’Bryan
January 20, 2018 1:27 pm

“but that doesn’t cover the useless nature of the results”

What else needs to be said?

January 20, 2018 1:41 pm

I will step into the deep end of the pool: –
From Dr. Jaworoski’s article linked in Dr Ball’s paper. Not written “(1992!)” to quote from a commentator above in this thread, but rather published in Science March 16th, 2007, giving Dr. Jaworoski some 15 years to repent and correct any of his insights in his earlier work before commenting on the (“corrupted”, appropriated with apologies to Dr Ball) state of Climate Science today – {given the reader will accept that since 2007, a multiple array of falsifications of the CAGW postulate have only further invalidated the “theory”, if one can call it that…and that in the ensuing last decade, no sign of corroborating ‘proofs’ have emerged from the CAWG billions$ – nor, of course, any sign of ‘tropical hot spot’, SLR, ‘acidification’ thereof etc etc ad infinitum)!
From the article – page #5 in the pdf below the photo, page 42 in the Science issue: –

“This is the reason that between 1880 and 1940, when the global average temperature warmed up by about 0.5°C, the direct measurements in the atmosphere registered a very large increase of CO2, from about 290 ppmv in 1885 up to 440 ppmv in 1940—about 60 ppmv higher than now (Beck 2007).”

While I disabused myself of any belief in this monstrous, murderous cult in the mid-90’s, hell-spawn afterbirth of Maurice Strong and his Globalist Un-Kosher Cabal (read: – i.e. Club of Rome, plus the Usual Suspect Malthusian eugenicists), and quit canvassing for Greenpeace (well, I admit the precipitant was a very unpleasant debate with another acolyte wherein I was deemed a “specie-ist” for favouring the survival of Dolphins OVER Tuna…and saw “the writing on the wall” to borrow a Biblical allusion), in all my subsequent explorations and in large part due the incredible work of and education I absorbed from John Daly (RIP) and his “Still Waiting For the Greenhouse” site, I never came across this seminal factoid – that CO2 ppmv reached 440 in 1940…!!! – Which duh!, subsequently declined. Makes sense of course, given the N Am (and wider geographic?) heat of the mid-to-late 30’s…

‘Nuff about me. Felt I had to share. I’ll make place for those inclined to math, physics and statistics.
Does occur to me that the fact-resistant among us might respond a bit more to ‘dumbing down the social conversation’ with factoids like that above.

If the mods want to blockquote what I excerpted from Dr Jaworoski above, have at it…I’ll take the learning curve on that, eventually…

Best and respectfully to all

ps: – Is it my imagination, or is it not Dr Jaworoski, whose work with radioactives in the environment (i.e. Chernobyl) is apparently lauded and well-respected by the commentator aforementioned above, who applied his skills in radioactive isotopic analysis of solar-driven temperature fluctuations evidenced in the ice records to predict what may be far more than merely a “Maunder Minimum” (200/400yr) equivalent, upon we all, as write to you, my dear readers? Which, IMHO, is a very valid candidate for application of the “Precautionary Principal”?

[Rolf Witzsche really ought to get a narrator for his prolific video productions on YouTube, his accent and proclivity for conflating (?)(perhaps!) religious and Xtian Biblical aspects of the challenge he postulates we face, do require a degree of patience to see beyond…- ]

pps: – But of course, that would mean a little ‘adjustment’ in social focus by the MSM and their Progressive Left Masters. Never Mind.
Perhaps, after the USA mid-terms, and we’ll see if The Donald can carry his social media Genius into another term, in face of the Deep State. Hey! Release the Memo# ;-D

Reply to  Tiburon
January 20, 2018 2:08 pm


All the objections from Jaworowski were written in 1992:
aworowski, Z., T.V. Segalstad, and N. Ono, Do glaciers tell a true atmospheric CO2 story? The Science of the Total Environment, 1992. 114: p. 227-284.
Alle what he has done is repeating the same arguments in 2007, as if het didn’t read the work of Etheridge e.a. from 1996 or any other work about ice cores. Just sad.

About the work of the late Ernst Beck: I had several direct discussions with him when he still was alive. The problem is that he lumped everything together: every known measurement of CO2: the good, the bad and the ugly. The average of any year was the “background” CO2 level of that year, according to him. Even if at one end of the world the result was 200 ppmv and on another point 500 ppmv. Zero quality control, no removing of unreliable data (like measuring under, in between and above growing crops…).
See further:

Reply to  Ferdinand Engelbeen
January 20, 2018 4:18 pm

Thank you for your very civil reply and for the information. I have no pretensions to anything approximating a ‘science background’…but do enjoy research and we are blessed to have both Wiki and Google (and dictionaries! LOL) to assist in such interests. I will explore as you suggest.

I’ll accept at face value in the interim your analysis of Jaworoski’s response to the latter work in regards ice core. I’ll say as a lay person that without the dogged and intense work you clearly do ‘sifting’ the core data and examining thereupon the assumptions that are the shaky understructure of warmists’ rejection of all but the demon gas as being a ‘thermostat’ for our world’s temperature, we … deplorables…would have no choice but to reject the evidence of our lying eyes. For that, sincere thanks.

I believe (that’s it, no evidence), that something else is on its’ way, on multiple fronts. I’ve come to believe for decades now, that we live in A World of Lies, and be it politically, economically, socially, and yes, spiritually, we are very much challenged with ferreting out a semblance of comprehensive Truth.

That said, I also see very much to celebrate, in your work and obvious attention to detail of fact and data, and therefore not least,, within the …Zeitgeist.
But of late I’ve been investigating the very medium with which we reach each other in this forum, and ‘testing’ search engines, speed of propagation of memes and transparency of news distribution, and what I’m finding is, euphemistically-speaking, disturbing.
Here on this site there is an incredible wealth of specialized, discerning expertise, but alas, I also see little in the way of generalization (let us say apart from Anthony Watts himself) that can stand against the juggernaut of the AlGore and his media (w)hore-des. We shall see.
Wouldn’t it be a bitter irony if Climate in fact DOES catch up with us?
Don’t sell your fur coats and hats. And keep watching the skies. And if so inclined, I for one accept that Prayer…works.

Reply to  Ferdinand Engelbeen
January 20, 2018 4:20 pm

I was unaware of Ernst Beck – and his work. Have Wiki’d, and am intrigued, Thank You.

Reply to  Tiburon
January 20, 2018 2:50 pm

Plus, the beck data correlates a change in temperature with a change in CO2 which is known to be false in the modern (MLO) record. If there is a temperature relationship in the modern record, it is that the CO2 growthrate correlates with temperature. (recent data falsifies the beck data)…

Reply to  afonzarelli
January 20, 2018 3:11 pm


~graph courtesy of F.E. It appears to be a little off. The increase in the growthrate from around the turn of the century to 1940 should not be that steep. (more like 1 ppm/year change as opposed to 2 ppm) The difference between the temp calc and ice cores is more like 10ppm max which falsifies beck.

ferdinand, i can’t actually see this graph as i write, so i’m winging it from memory…

Reply to  Tiburon
January 21, 2018 10:01 am

Tib- didn’t see the pdf link referenced. can you post? (or did i miss?)

“From the article – page #5 in the pdf below the photo, page 42 in the Science issue: –”””

January 20, 2018 1:50 pm

Hey, you guys can test the CO2 levels in the ice cubes in my freezer any time you wish.

This is an informative article, although I’m not sure what I’m supposed to do with the information. My real concern is how much money is wasted trying to settle what seems to rapidly be turning into an argument about “i’m right – you’re wrong” from various sides of the table, and it almost always centers around cash flow.

Aside from some undefined need (other than ego feed) and the politics that are driving the arguments now, I don’t much see the point. What happened 1,500,000 to 600,000 years ago is going to happen again, and there is NOTHING – ZERO, ZIP, NADA – that we can do about it other than be prepared for it. And you don’t do that by threats and political intimidation, or corrupt grabs for money.

Ice ages are a recurring cycle on this planet now and will continue to be that way until plate tectonics change the surface formation, which will ALSO happen. There is nothing any of these people can do to stop that, or any future ice ages. PERIOD.

I admit that I would enjoy seeing Mann running from a southbound glacier, but that’s a cartoon, so I’ll let it go.

Reply to  Sara
January 20, 2018 2:00 pm

but but Sara……when CO2 levels got above…..well above something in the past….it turned into fireball earth…..no wait…that was snowball earth…..well anyway, it doesn’t matter, we can’t let CO2 get above whatever that was

(I ate 3 fudge brownies and can’t be held responsible for anything I say right now)

Reply to  Latitude
January 20, 2018 2:23 pm

Latitude, I respect your honesty about the brownies, but I have chocolate ice cream that has enough fat in it to stun a whale. I cannot be held responsible for anything I say after I’ve indulged in that ice cream, either.

I think we all need to keep up the dialog about how lucky we humans are to live on the only truly habitable planet that we know about [so far], and how difficult it may be to find another planet to expand to, never mind get there. The Goldilocks zone doesn’t do so well when all planets in a system are at a distance inside Earth’s orbit from their central star. We’re luck to be here.

Logic says there must be some reason for our existence beyond this silly squabbling over technical terms, etc. But maybe I can simply see further than the people who think we’re all doomed and want our money and ice cream.

Reply to  Latitude
January 21, 2018 12:32 pm

The world’s climate was perfect on June 6, 1850.

It’s been downhill since then.

Might as well eat all the fudge brownies.
Global warming will soon end life on earth
as we know it — so what difference will it
make if you weigh 375 pounds?

Reply to  Sara
January 20, 2018 3:22 pm

Sara and Latitude, But wait the average temperatures in 1972 is the ideal temperature for the world to stay at. So we must do everything in our power to maintain that temperature. Especially spending large amounts of money to keep it that way. Al Gore says so. Uh oh I just had moose tracks with brownies ice cream.

John Harmsworth
Reply to  garywgrubbs
January 20, 2018 4:09 pm

I think the ideal temp got changed to 1872 equivalent or thereabouts. So humans can be blamed even if they’re dead.

Reply to  garywgrubbs
January 20, 2018 5:39 pm

Noted, garywgrubbs.

And my response is that, if Algore and others of his ilk, and all those who agree with that philosophy, would consider getting their mouths sewn shut, not only would the CO2 levels be stabilized for the foreseeable future, but it would also free up massive resources in the way of food, electricity production, and gas for heating and cooking, never mind extensive loads of petroleum to refine for vehicle use and synthetics production, so that the rest of us could live our lives in comfot, knowing that the sacrifices of those CAGWers, Warmians, and Greenbeans were not in vain.

I think they should be willing to make that sacrifice just to save the planet, don’t you? We could build a monument to them: a stone circle perhaps, with 56 pillars carved with the phases of the moon. Each cycle of the moon would be an opportunity to study the sky for timing of agricultural endeavors and/or hunting, herding, planting,lambing, calving, and harvesting. People from all corners of the land would get together, peacefully assembled, to laud these legendary souls who willingly sacrificed everything to save the planet.

They will be missed. Or not.

I think turtle cheesecake with the chocolate ice cream is in order tonight….

January 20, 2018 1:55 pm

I appreciate skeptical skepticism. But let me naively ask the obvious question about what then do those slowly changing millennial gas measurements reflect? They are not random measurement results and there is a consistency in their slow and constant change. What is that?

Reply to  Arbeegee
January 20, 2018 2:41 pm


Over the lifetime of the earth, CO2 levels have been much higher than today. Despite that, we have had colder and warmer temperatures. Nowadays, most of that CO2 resides in thick chalk layers all over the world and we are at a quite low CO2 level, compared to millions of years ago.
What we see in the ice cores is that over the past few millions of years, CO2 follows temperatures with some lag (800 +/- 600 years), probably a matter of deep ocean – atmosphere circulation and changing temperatures, while the opposite (the influence of CO2 on temperature) is not measurable.

Since about 1850, the main start of the industrial revolution, CO2 levels in the atmosphere increased about in ratio with human emissions of CO2. It is as good as certain that human emissions caused that increase, which benefits near all plants on earth.
(If you are interested, here are the arguments:
http://www.ferdinand-engelbeen.be/klimaat/co2_origin.html )

Some sceptics here don’t accept that humans are the cause of the CO2 increase and all kinds of arguments are used, probably because if humans aren’t to blame, the whole AGW meme is gone, as that is one of the pilars of the theory.
Unfortunately, that humans are the cause of the increase is rock solid and the only point where the “warmistas” are right and the sceptics are wrong. Using arguments which don’t hold to any scrutiny is shooting in your own foot and weakens your arguments where the other side has little defence, like the (lack of) influence of CO2 on temperature…

The above writing from Dr. Ball is a clear example of using the wrong arguments…

Reply to  Ferdinand Engelbeen
January 20, 2018 6:21 pm

As a scientist but not in the areas needed to cover global warming/climate change, I keep your site on my desktop!
You explain the science extremely well. Easy to follow and yoru documentation, everything with appropriate references.
I agree with your Conclusions:

“As shown in chapter 1, there is little doubt that humans are fully responsible for most of the increase of CO2 in the past (at least halve) century, that means that – as far as there is an influence of CO2 on temperature – that humans may be responsible for (a part of) the temperature increase. How much, that is an entirely different question, as that mainly depends of the (positive and negative) feedback’s that follows any increase of temperature…”

My only question to you would be to look at the recent El Ninio years (1998 and 2016) for the data on CO2 at Mauna Loa. I see that in these years, the peak CO2 is higher than the corresponding previous years and then, at least so far we can see the return in years after 1998 but to early to tell after 2016. This to me support your above conclusions that humans are fully responsible for “most of the CO2 increase in air” but from time to time mother nature must participate in the increase of CO2 in air from oceans releases of CO2. Now the sources of this CO2 being release from warmer oceans in these years is intriguing.

Reply to  Ferdinand Engelbeen
January 20, 2018 7:06 pm

I don’t recall seeing on this site the claim that increasing CO2 was NOT caused by man, except indirectly as CO2 follows temperature so it MAY BE naturally increasing. There is a George Carlin skit on CAGW. Towards the end, he posits that maybe Gaia created us to invent plastic, because there wasn’t any. You will notice in long term CO2 reconstructions, CO2 levels dropped to some 180 PPM during the last two glacial periods. At 150 PPM most food crops starve so we would starve. Do you suppose our purpose is to release sequestered CO2 in order to save all life on earth??

Reply to  Ferdinand Engelbeen
January 20, 2018 9:26 pm

“…that humans are the cause of the increase is rock solid…”
Maybe but when you actually see how little CO2 is released by humans compared to natural phenomena, the unknown and error factors involved in most all climate measurements, and the seemingly continuous discoveries of new and unexpected sources of warming gases, and solar factors not considered or well understood, combined with the benefits of a CO2 rich Earth, that solid rock begins to look more and more like soapstone.

Reply to  Ferdinand Engelbeen
January 21, 2018 2:51 am

“What we see in the ice cores is that over the past few millions of years, CO2 follows temperatures with some lag (800 +/- 600 years), ”

Unfortunately we only have data going back c. 795,000 years, and that often quoted 800-year delay figure only applies to terminations i. e. the very fast temperature rise at the end of glaciations. During the cooling phase at the start of glaciations the lag is much larger, on the order of 5,000 years.
This is particularly obvious during the early phase of the last glaciation (MIS 5a-5d, 75-115,000 years ago) with two warm and two cold phases. The temperature swings were so fast that at times temperatures and pCO2 were going in opposite directions for thousands of years.

Since the CO2 presumably comes out of and goes back into the oceans the most likely explanation is that during terminations ocean and land temperatures track each other while during cooling the oceans lag a lot (the 800 year lag during warming is almost exactly the inevitable delay that you would expect from the turnaround time of the thermohaline circulation, the deep ocean simply can’t react faster than that).

It would be extremely interesting to have ice cores going back a few hundred thousand years more and to be able to see how pCO2 reacted in the “41 kyr-world” with much smaller and faster glacial cycles.

Reply to  Ferdinand Engelbeen
January 21, 2018 3:35 am


Thanks for your kind words…

About the CO2 rate of change: while in average the CO2 increase in the atmosphere is 50-55% of human emissions (4.5 GtC/year – 2.15 ppmv from ~9 GtC/year human emissions), there is a huge variability in the year by year CO2 rate of change: more in warm (El Niño) years, less in Pinatubo and other cool years. In general that levels out around the average increase within a few years. While the obvious culprit is ocean temperatures, the highest influence is from tropical forests which suffer drought and too high temperatures, thus less uptake and still maximum decay, during an El Niño and the opposite later on plus extra uptake of CO2 after the Pinatubo eruption as the light scattering from the stratospheric dust increased photosynthesis of leaves normally in the shadow of other leaves during part of the day.
How do we know? If CO2 changes are mainly by the oceans, CO2 and δ13C changes parallel each other (ocean CO2 has a higher 13C level than the atmosphere). If mainly by vegetation, CO2 and δ13C changes oppose each other (CO2 from decomposing vegetation has a lower 13C level). The latter is the case:


where CO2 rate of change changes lag temperature rate of change changes with about half a year and δ13C rate of change changes are synchronised with CO2.

The variability of the extremes (El Niño, Pinatubo) is about +/- 1.5 ppmv in any year around the 90+ ppmv trend over the past 60 years.
Temperature variability is the main cause of the year by year variability of the CO2 rate of change, but temperature is not the main cause of the increase itself. The natural (dynamic) equilibrium per Henry’s law would be about 290 ppmv for the current average ocean surface temperature with a change of ~16 ppmv/K or about +13 ppmv since the LIA due to the temperature increase. The rest is by humans…

Reply to  Ferdinand Engelbeen
January 21, 2018 5:47 am


“…that humans are the cause of the increase is rock solid…”
Maybe but when you actually see how little CO2 is released by humans compared to natural phenomena

Most of the natural carbon cycle is seasonal: oceans and vegetation exchange a lot of CO2 (+ and – 150 GtC/season) in and out (opposite to each other), but that levels out to near zero after a full year. The observed year by year natural variability is +/- 3 GtC (+/- 1.5 ppmv) after full years.
Humans add about 9 GtC (4.5 ppmv) per year of which about half remains in the atmosphere. That is over twice the natural variability and twice the observed increase in the atmosphere. Thus while the human contribution is relative small, the natural contribution at the end of the year is negative: nature is a net sink for CO2…

Reply to  Ferdinand Engelbeen
January 21, 2018 7:17 pm

It’s complete bunk. The rate of change is proportional to appropriately baselined temperature anomaly, which puts the arrow of causality firmly in the direction of temperature to CO2, and not the reverse:


If you give me the starting point, and the temperature history, I can reconstruct CO2 level. Human inputs need not apply, because their impact is negligible. Ferdinand’s kluge to add in human inputs to a high pass filtered version of the temperature relationship is physically impossible.

There is no doubt about it. Eventually, the truth will out.

Reply to  Ferdinand Engelbeen
January 22, 2018 3:58 am


As usual…

Your theory is mathematically possible for the past 60 years, but that is all. For the rest, it violates every single observation: Henry’s law for the solubility of CO2 in seawater (no matter static or dynamic), the δ13C decline, the δ14C decline, the process characteristics, the mass balance,…

That humans are responsible for the increase fits all observations.

But we can repeat the same again and again without effect. So let’s discuss ice cores, not any other points of discussion here…

Reply to  Ferdinand Engelbeen
January 22, 2018 4:42 am

Ferdinand Engelbeen, thankyou for your comments. You explain a complex subject with great clarity and patience.

Reply to  Ferdinand Engelbeen
January 22, 2018 5:47 am

The last 60 years saw the concentration go from 315 ppm to over 400 ppm. That is the era of the major observed rise. What happened before that is conjectural, due to the lack of reliable data, but it stretches credulity to imagine that the dynamic underwent wholesale change at that time.

Henry’s law is completely compatible with the observations in a dynamic framework, and the isotope results are conjectural.

Reply to  Ferdinand Engelbeen
January 22, 2018 7:09 am


I will bite with only one question:

– The variability in rate of change is caused by the temperature rate of change which drives the CO2 uptake/release of mainly vegetation (tropical) forests. That is observed.
– The slope in rate of change is NOT from vegetation, as that is a net, increasing sink for CO2. That is observed.
– Why do you insist that variability and slope are from the same process, as you use one offset and factor to match variability and slope?

Reply to  Ferdinand Engelbeen
January 22, 2018 6:54 pm

It only takes the scale factor to match the variability and the slope in the rate of change. The offset does not affect either. The temperature baseline is arbitrary, so there is no basis for criticizing the offset.

Reply to  Ferdinand Engelbeen
January 23, 2018 1:52 am


It only takes the scale factor to match the variability and the slope in the rate of change. The offset does not affect either. The temperature baseline is arbitrary, so there is no basis for criticizing the offset.

No matter the offset, you use one factor that should cover both the amplitude and the slope. Depending of which dataset you use, you can perfectly match either the slope or the amplitudes and only in rare occasions both with the same factor. That is completely spurious as both are the result of different causes.

Reply to  Ferdinand Engelbeen
January 23, 2018 6:25 pm

“…you use one factor that should cover both the amplitude and the slope.”

And, it does. Both the trend and the variability in the rate of change are matched by the single scale factor.

Reply to  Ferdinand Engelbeen
January 24, 2018 6:13 am


It doesn’t if you take the full Mauna Loa CO2 data series and the HadCRU4 temperature series. By using the satellite era you don’t include part of the 1945-1975 period where temperature went down while CO2 went up.
Either you match the slopes or you match the amplitudes, not both at the same time.

Reply to  Ferdinand Engelbeen
January 24, 2018 7:30 am

Mods: sorry wrong end tag closure in the above…

Reply to  Ferdinand Engelbeen
January 24, 2018 11:40 pm

Ridiculous. Even the temperature measurements themselves do not match one another. Does that mean they’re all invalid?

These are stochastic data, not deterministic. The match is as good as one gets in the real world.

Reply to  Ferdinand Engelbeen
January 25, 2018 12:36 am


Satellites and ground stations don’t measure the same air column, thus show a slightly different slope. If you align the slopes, the amplitudes match each other.

That is not the case for the match of T and dCO2/dt, as these are from different processes. That is physically proven as most of the variability is caused by vegetation and vegetation is not the cause of the CO2 increase in the atmosphere. Vegetation is a proven, increasing sink for CO2.

But I know, you don’t accept any observation that may even remotely oppose your theory…

Reply to  Ferdinand Engelbeen
January 26, 2018 10:01 pm

It is depressing thinking of how many people you probably take in with such rationalizations.

January 20, 2018 1:59 pm

There are certain word or ideas that mark out climate ‘science’ and one of them is without doubt ‘better than nothing ‘ by which proxies are used because actual measurements of elements are impossible to find. From magic tree rings to bubbles in ice they are used not because they offer good and valid data but because they are ‘better then nothing ‘
Indeed this weakness is actual a ‘useful’ factor , for it allows the opportunity to ‘adjust’ and the use of ‘statistics’ to generate desired result in the knowledge that is not possible prove these approaches invalid based on how they have changed previous ‘correct results’

And on this quicksand they have built ‘settled science ‘ !

Reply to  knr
January 21, 2018 11:48 am

The people who make the scary predictions …

own the average temperature calculations / data.

They predict something they control.


Extreme Hiatus
January 20, 2018 2:20 pm

“What do the Ice Core Bubbles Really Tell Us?”

That the ‘science’ is NOT settled.

Reply to  Extreme Hiatus
January 20, 2018 2:25 pm

What do the ice core bubbles really tell us? That the water wasn’t warm when it froze, snowflakes are white because they’re mostly air, and that there have been prolonged periods of cold, punctuated by rather short periods (by comparison) of warmth.

Reply to  Extreme Hiatus
January 21, 2018 11:46 am

“Modern climate science” is unsettling.

January 20, 2018 3:01 pm

Ferdinand, I agree with your analysis except for one major issue. This is the bacterial use of CO2 as a source of C for reproduction, thus removing CO2 from the ice and or bubbles. I do not know if the amount removed is significant, but recent work (and I do not have the sources handy) has shown the surprising fact of bacterial activity in ice layers. As the level of CO2 would drop, it would become more difficult to sustain bacterial activity,so the value would tend to level off.

Crispin in Waterloo but really in Ulaanbaatar
Reply to  Leonard Weinstein
January 21, 2018 12:31 am


You support my note to Ferdinand above. You are correct: there is water within the ice and bacteria live in it. We know there are sulphur and CO2-consuming life forms. Their influence in unquantified.

January 20, 2018 3:10 pm

I know this doesn’t apply exactly to Dr. Ball’s article. Today I was studying Barrow Alaska’s temperature records for December, January , February back in the 1920’s and 1930’s when they first started recording.
There are many highs in the 40s and a number in the 50’s during these years. Many more than recent years.
I am a lawyer and not a scientist, just a climate enthusiast. This clearly indicates to me that there could not have possibly been winter ice near the north coast of Alaska in those years. It would therefore make sense that winter Arctic ice cover was much lower and the globe much warmer in those years. Any feedback.

Reply to  mark
January 20, 2018 7:52 pm

My grandfather went to the Klondike during the gold rush. He was there from 1898 to 1902. I have his journal. He recorded everything, including temperatures. I haven’t finished transcribing the text yet, but he does record a wintertime temperature indoors in the cabin he and three friend occupied near Whitehorse as -6F. If I find any other such reports, I will bring them up.

Crispin in Waterloo but really in Ulaanbaatar
Reply to  Sara
January 21, 2018 12:38 am

Hey! The high today in Ulaanbaatar is -28 C. Please don’t tell me it is colder here now than Whitehorse was in 1900!

The wind chill forecast for tonight has been set at (“feels like”) -55 C. I think I will step outside just to see my eyelashes get covered with frost. That’s worth a selfie.

It was -67 C (without wind chill) in Yakutia last Tuesday. I think the Yukon has it easy.


January 20, 2018 3:11 pm

Ferdinand, search “Bacterial activity in ice”. In particular look at V. Mateva 2008, and Scientific American blogs for Oct. 10, 2013. It is clearly shown that activity persists at even -40C.

Reply to  Leonard Weinstein
January 20, 2018 3:53 pm


Bacterial activity depends of two items: temperature and food. In the case of Antarctic ice cores, baterial inclusions are mainly found around dust inclusions, which are in general at maximum at the depth of the glacial periods, just before the start towards a new interglacial.
At the -40ºC of the Vostok ice core, most biological activity is constrained to survival by DNA damage repair. That needs energy, mostly found in oxydising ammonia to N2O and using CO2 as carbon source for synthesis. If we assume that all N2O produced was made that way, that repair process would have consumed 0.5 ppmv CO2 at places of high dust inclusion. Not really a problem.

See: http://www.pnas.org/content/101/13/4631.full.pdf
Item K. is specific for the Vostok ice core.

January 20, 2018 3:29 pm

It’s just a statistically averaged proxy based on glacial ice that ignores the realities of statistical error like every other proxy on Earth with the exception of sea floor mud deposition of invertebrates.

Carbon Dioxide, Oxygen, Hydrogen, Nitrogen, Helium and Fluorine can all migrate through crystalized ice when placed under pressure and all do so at different rates.

Baring the aboved mention of bacterial growth based on “entombed” CO2, you also have to consider the electrostatics of Glacial environments and accept that some of the water Hydrogen sublimes out of the ice and is resultingly replaced by Carboxyls which have never been tested for because nobody keeps the goddamned cores under pressure during extraction, storage, shipping and testing..

Ice cores are the LEAST viable of all proxies.

Reply to  prjindigo
January 20, 2018 4:16 pm


Ice core gas composition is not a “proxy”, but direct measurements of ancient air, be it a mix of several to many hundreds of years.

These gases may be pressed through ice if the pressure difference is high enough, but there is not much pressure difference between 2000 and 2001 meters of ice…

There is no measurable migration of CO2 or any of the other gases through the ice matrix in ice cores. There was an attempt to calculate a theoretical migration speed by measuring the increase of CO2 near melt layers. That shows a broadening of the resolution from 20 to 22 years at middle depth and to 40 years at full depth for the relative “warm” (-22ºC) Siple Dome ice core. At -40ºC in the Vostok or Dome C ice core, not much (theoretical) migration is left.

There is some “last minute” migration of the smallest molecules/atoms like O2, Ne, Ar and others when the pores reach a minimum diameter during bubble closing and again during expansion/storage. CO2 is not one of them:

Bacteria in inland Antarctica ice is no problem at all.

Reply to  Ferdinand Engelbeen
January 21, 2018 7:20 pm

This is all hand waving. It’s just saying what you think, with no physical evidence backing it up.

Reply to  Ferdinand Engelbeen
January 22, 2018 3:49 am


The theoretical migration of CO2 was measured near melt layers in a relative warm coastal ice core and the CO2/T ratio in inland ice cores was measured over the past 800,000 years. No decrease of that ratio over 8 interglacial, thus no measurable migration. Do you have any proof of the opposite?

Reply to  Ferdinand Engelbeen
January 22, 2018 5:49 am

You cannot run the experiment from start to finish. You can only speculate on what you think the end result should be. That leaves it all unverified and unverifiable.

Reply to  Ferdinand Engelbeen
January 22, 2018 7:27 am


One can discuss the problems with ice cores at length and pose lots of objections, but as long as you don’t give any proven physical reason why the ice core data must be wrong, at least one can accept the data as they are available as a working hypothesis.

What is proven:
– CO2 at closing depth is the same in still open pores as in already closed bubbles.
– CO2 in the ice bubbles and in the atmosphere is the same for at least the period 1958-1978.
– CO2 in ice cores with extreme differences in temperature and accumulation rate is the same over all overlapping time spans for the same average gas age.
– Theoretical migration of CO2 in coastal ice cores is small to modest.
– There is no measurable migration in much colder inland ice cores.

If you have no proof that one of the above is false, or any proof that ice cores CO2 is unreliable, I don’t understand why you object to using the ice core data at least as an indication of the past CO2 levels.

Richard G.
Reply to  Ferdinand Engelbeen
January 22, 2018 11:12 am

Ferdinand, proxy evidence is conjecture. Even if well reasoned it is conjecture.
copied from up thread:
To add a pinch more uncertainty to the witches brew of conjecture that is proxy evidence, consider the phase change diagram for CO2: gas to solid (precipitate) at -78Deg C at one atmosphere pressure. This presents the possibility of there being a CO2 fraction within the snow that falls under extreme cold conditions.
“The lowest natural temperature ever directly recorded at ground level on Earth is −89.2 °C (−128.6 °F; 184.0 K) at the Soviet Vostok Station in Antarctica on July 21, 1983 by ground measurements.
A 2009 study estimated that under exceptional climate conditions similar to those recorded at Vostok 1983, temperatures higher on the plateau around Dome Argus could drop lower than −95 °C (−139 °F; 178 K).[1] On August 10, 2010, satellite observations measured a surface temperature of −93.2 °C (−135.8 °F; 180.0 K) at 81.8°S 59.3°E, along a ridge between Dome Argus and Dome Fuji…” https://en.wikipedia.org/wiki/Lowest_temperature_recorded_on_Earth#cite_note-2

Reply to  Ferdinand Engelbeen
January 22, 2018 11:39 am

Richard G.,

Copied from above:

The partial pressure of CO2 is 0.0004 bar in the atmosphere. With 1 bar it may precipitate at -78ºC, but you need much lower temperatures to condense any CO2 at 0.0004 bar pressure…
Neither does 1% water vapor in the atmosphere precipitate at 100ºC, but it does if it is 100% water at 1 bar…

Richard G.
Reply to  Ferdinand Engelbeen
January 22, 2018 12:03 pm

Not being a Psychrometrist, perhaps you could tell me what the partial pressure of water vapor and relative humidity would be at <-78C. Dew point would also be interesting.

Reply to  Ferdinand Engelbeen
January 23, 2018 1:30 am

Richard G.


Water vapor at -73ºC (200 K) is about 1 μbar according to that diagram, at lower temperatures thus below 1 μbar.
If the air started with any water vapor pressure above 1 μbar and was cooling down, the air get saturated and the excess water vapor above saturation freezes out.
From the moment on that the temperature reaches the original dewpoint of the air/vapor mixture, the dew point remains the same as the temperature with 100% RH.

January 20, 2018 4:29 pm

When “climate scientists” said ice would disappear?

USS Little Rock (LCS-9), the latest Freedom-variant littoral combat ship to join the fleet, is iced in on the St. Lawrence Seaway in Montreal and doesn’t plan to move until the Spring thaw.


….this news yill not appear in NYT, CNN etc…

Reply to  AlexS
January 20, 2018 7:28 pm

I appreciate that President Trump resuscitated a moribund economy, brought both Hispanic and black unemployment down as low as they have ever been since they have been measured, got the new unemployment claims down to a 40 plus low, and brought North Korea to the negotiating table. He also solved global warming. But…….did he have to overdo it??

Reply to  Jon Jewett
January 20, 2018 7:53 pm

Is that why Chuckie Shumer is mad at him?

January 20, 2018 4:48 pm

Likely typ here.

“How is that useful for climate when a single sample for the entire period of the Holocene Optimum.”

May be…

“How is that useful for climate when a single sample CAN REPRESENT the entire period of the Holocene Optimum.”

January 20, 2018 5:40 pm

Just to add another alarming and also amazing study about bubbles in ice cores.
Seems CO2 data from ice cores are totally crap.

“Do glaciers tell a true atmospheric CO2 story?” (Zbigniew Jaworowski)

Reply to  petermue
January 20, 2018 11:59 pm


Completely outdated and completely wrong: rebuted already in 1996 by the work of Etheridge e.a. on three ice cores at Law Dome and Dr. Jaworowski made such physically impossible errors that all what he said should be looked at with a lot of scepsis. See:

January 20, 2018 5:44 pm

I think Stomata data would be a far better proxy for CO2.

Reply to  J.H.
January 21, 2018 12:05 am


By far not at all: ice cores bubble contain the real composition of ancient air, but a mix from 10-600 years, depending of how fast the bubbles are fully closed, which depends of the local snow accumulation. The composition (including isotopes) is measured with the same equipment as for direct air measurements. Accuracy +/- 1.2 ppmv for the same ice core, +/- 5 ppmv between different ice cores for the same average gas age.

Stomata data are proxies, influenced by a lot of other items than CO2 levels and by far less accurate.

Clyde Spencer
January 20, 2018 6:22 pm

I was drafted into the Army in January 1966. After Basic Training, I was assigned to the Cold Regions Research and Engineering Laboratory (Hanover, NH) as a Physical Scientist, as part of the Scientific and Engineering Personnel Program. That Summer, I was sent to Camp Tuto (Greenland), about 20 miles from Thule Airbase. I was there to supervise a closure (from plastic deformation) survey in an ice tunnel that was constructed a decade earlier. One of the distractions that we had to put up with was Danish pilots and stewardesses who came to the tunnel to collect ice for cocktails. One had a superb cross section of the ice on display. The tunnel was originally high enough to drive a 2 1/2 ton truck (aka Deuce and a Half) in! The ice notably alternated between clean layers suitable for addition to drinks, and very dirty layers, suitable for making moraines where they intersected the ablating snout. As I recollect, there were some of the clean layers that were preferred for sampling because they effervesced in the water-alcohol solution more than ice from the other layers.

A couple of points to be made. The ice layers were not horizontal; they were inclined. The glaciologist at the CRREL lab had named the features evident at the snout as shear moraines. I don’t think that you will find the term in any books on geology or glaciology. Basically, with an uneven bedrock topography, the ice shears up over an obstruction, with stagnant ice in the ‘shadow’ of the positive relief. (In New England I have observed glacial striations in the bedrock suggesting that the moving ice can also be deflected laterally.) If this happens numerous times, the stratigraphy can become quite complex and disordered. So, assigning relative age of ice layers near the base of glaciers can become problematic. If, indeed, the air entrained varies between layers, then someone needs to explain how that happens.

As is often the case in geology, things may be more complex than they appear at first blush.

Reply to  Clyde Spencer
January 22, 2018 11:58 am


The stratigraphy is one of the reasons that near all ice cores are drilled at the top of ice domes. Even so, the lowest layers may be disturbed in part by movements, in part by the higher temperatures due to earth warmth.

One exception is the Vostol ice core, which was more downslope and reached 420,000 years until a (large) part was disturbed ice as the glacier passed a ridge just upstream. Under the ice is a large lake, Lake Vostok that was reached a few years ago with a lot of precautions to prevent contamination with external bacteria.

January 20, 2018 8:27 pm

Just to pick up on 2 comments made above:

Water crystallised as snow and trapped air in the eventual ice are supposed to be an accurate representation of the oxygen isotopes and atmospheric gasses present at the time of snowfall/ice compaction, which other commentators suggest have an accuracy of 50 years. This time range is supposed to represent the range between the water vapour crystallising as snow through to firn compacting as ice.

Snowfall onto a stable polar ice dome say in Antarctica or Greenland is a very different situation to snowfall high in mountain ranges and then transporting laterally and downward in glaciers, and compacting to firn then ice as cover weight increases. This is why the polar ice cores are sighted on thick domes, often in the centre or in sub-basins.

January 20, 2018 8:58 pm

Interesting site. I started off being sceptical of ice cores but I finished up being sceptical of the author.

Isn’t Ferdinand Engelbeen (from above) correct? On timescales of decades at least, ice cores seem to be an invaluable proxy and tell us a great deal. Why try and cast doubt on them like that with the suspicious, 26 year old opinions of Dr. Jaworowski? Huh?

Reply to  zazove
January 21, 2018 3:33 am

I agree, but “centuries” or even “millenia” is more correct than “decades” except for the very recent past. Decade timescales are in practice only applicable to the Law Dome data. What is somewhat disquieting in this context is that even after 20 years the oxygen isotope data for most of the Holocene from the Law Dome have never been published. In particular since the Climategate emails shows that van Ommen was specifically warned against releasing the data to Steve McIntyre.

Reply to  tty
January 21, 2018 6:11 am


The longest (downslope) Law Dome ice core is only 2000 years (gas age), ice age may be a few hundred years longer, but I don’t think it will span the whole Holocene?
Further, I suppose that the Law Dome 18O data are a good indication of local temperatures, not of SH temperatures, as the water vapor catch area is the nearby Southern Ocean, compared to the inland cores which have a much wider catch area.

I did find some interesting older work on Law Dome before the main drillings by Etheridge e.a.. looking at the 18O data more near the surface:
And some (partial?) data seems available:

Reply to  tty
January 21, 2018 7:55 am

The section 9,000-19,000 years ago has been published:


Notice what they say about the core (van Ommen who has the data was a co-author):

“The core site (18), near the summit of LD (66°46 S, 112°48 E), is characterized by a high rate of accumulation (late Holocene average, 0.68 m ice equivalent per year) that results in an ice core with a highly
tapered time scale in which the Holocene represents some 93% of the ice thickness of 1200 m. However, the full LD isotopic record generally matches the long records from Vostok and Byrd to at least 80 ka, indicating that the record is continuous and undisturbed over this period”

So, apparently continuous d18O data back to at least MIS 4/5a existed by 2002 (“the full LD isotopic record”). The Holocene data due to the exceptional accumulation would seem to be very interesting. The summit core was drilled 1987-93, but is still mostly unpublished 25 years later. Why?

Given the Climategate email one might suppose that somehow it would not help “the cause”:

“Hi Phil, Personally, I wouldn’t send him [McIntyre] anything. I have no idea what he’s up to, but you can be sure it falls into the “no good” category.”

Reply to  tty
January 21, 2018 12:44 pm

Thanks tty,

Indeed it seems that they have ice data for at least 19 kyear from the slightly downslope DSS core at Law Dome…

Reply to  zazove
January 21, 2018 7:22 pm

How can we know? How can we perform an end-to-end verification?

We can’t. We can only extrapolate from lab experiments, and make educated guesses. You delude yourself if you think they are anything more solid than just that.

Reply to  Bartemis
January 22, 2018 2:59 am


The problem with ice cores CO2 is that CO2 proxies all are inferior to the accuracy of the ice core measurements. The same problem we have had with atmospheric measurements with wet chemical methods in the pre-Mauna Loa past: +/- 10 ppmv, compared to +/-0.2 ppmv for the current NDIR method.

All we have is a small – 20 years – overlap with direct atmospheric measurements and “bootstrapped” overlaps between ice cores with extreme differences in temperature and accumulation rate. As these show about the same (+/- 5 ppmv) CO2 levels, there is little doubt about the conservation of CO2 and other trace gases in the air enclosed in the ice cores, independent of local influences.
Further, the fact that CO2 follows temperature changes in exact the same ratio, with a lag, over the past 800,000 years shows that both are directly linked and that there is no measurable migration of CO2 in the ice, or the ratio would decrease over time.

Besides that, proxies like in ocean sediments (foramins) show similar glacial-interglacial CO2 variability over longer time spans (up to a few million years), but less accurate.

January 20, 2018 10:17 pm

The message here seems to be that ice-cores are pretty unreliable. But the claim that temperatures started rising before CO2 started rising seems to be based on ice-cores. Is there good reason for believing this rather than, say, CO2 rises before temperature, CO2 and temp rise at the same time, or no correlation at all between CO2 and temp?

Reply to  RoHa
January 21, 2018 6:27 am


This is typical for the confirmation bias that plagues people in general, warmistas and sceptics alike…

If the data don’t confirm one’s opinions, the data must be wrong. If the data do confirm one’s opinions, they are accepted without any reservation…

Gas data from ice cores, including CO2, are quite reliable, but are smoothed over one decade to several centuries, depending of the local snow accumulation rate.

Calculation of the difference between the age of the ice and the average age of the enclosed gas bubbles is not that simple, as that depends of the above accumulation rate and the local temperature, That provides the time needed to fully close all air bubbles in the ice. For that purpose one has developed firn densification models, but even so in other periods the accumulation rate and temperature may have been much lower, or slightly higher.
Thus the lead or lag between temperature and CO2 levels is far more uncertain than the level of CO2 at a certain depth (even if the exact dating at that depth may be wrong).
Despite that, the general opinion in the ice cores community is that CO2 probably lags the transition between cold and warm periods with about 800 years and certainly lags the opposite transition with several thousands of years.

Reply to  Ferdinand Engelbeen
January 21, 2018 3:12 pm

So it’s “the general opinion” that CO2 “probably” lags rise in temp by 800 years. Thanks. That isn’t quite as definitive as I would have hoped, but still on the side I wanted.

(And of course the only data that counts is the data that supports my prejudices. Anything else must be wrong.)

Reply to  Ferdinand Engelbeen
January 21, 2018 5:06 pm

Ro, there is a school of thought out there that there is no lag of temperature behind that of CO2. What the lag may represent is the lag of global temps behind that of the temps at the poles. Shallow ice cores do not have an 800 year lag when compared with temperature reconstructions. If you think about it, it kind of makes sense. Coming out of a glacial, the temperature driven THC keeps temps cooler at the equator than they otherwise would be (and the round trip for the THC is 800 years). Conversely, the return into a glacial would have a much slower THC because, again, the THC is temperature driven. (cooler SSTs relative to the upwelling water causing weaker walker trade winds) That’s when the lag is in the thousands of years…

January 20, 2018 11:37 pm

Science is supposed to be about what can be observed and tested…IMHO ice cores aren’t quite the same thing. Too much is assumed or inferred.

Reply to  4TimesAYear
January 21, 2018 7:24 pm

Indeed. End-to-end validation is impossible.

January 21, 2018 2:31 am

In 2016 some WW2 fighter planes that had crash landed on Greenland were found below 264 feet of ice.
Did they sink through the ice or did the ice build up over them?
If the latter would the air bubbles near the planes be useable to cross calibrate the CO2 levels, as there must be some data on CO2 levels in 1942 when the planes crashed?

Reply to  StephenP
January 21, 2018 3:09 am

At that depth the bubbles have hardly even closed yet. However in maybe 50 more years there will be a sure-fire way to test how smoothed ice-core pCO2 values really are when “the bomb pulse” of increased C14 due the H-bomb tests in the 50’s and early 60’s has passed the bubble closure zone.

Reply to  tty
January 22, 2018 2:40 am


Bomb test 14C distribution is already measured in the Law Dome ice cores:
Unfortunately, the full article is behind a small ($ 6) paywall.

J Mac
Reply to  StephenP
January 21, 2018 6:46 pm

The ice built up over them, from pressure compacted yearly snow falls. The planes belly landed on the glacier in July 1942. Fifty years later (1992), the P-38 Lightning that became know as Glacier Girl was raised piece by piece from beneath 268 feet of compacted snow and ice.

Reply to  J Mac
January 22, 2018 3:05 am

Compacted snow (firn) turns into solid ice about 300 feet below the surface. It varies a bit depending on snow density, which in turn is determined largely by wind.

January 21, 2018 3:25 am

I don’t quite understand the motivation for this sweeping and unconvincing rejection by Dr Ball of palaeo reconstructed climate data. Is he a 6- day creationist? It does not help the climate skeptical position, of which the palaeo data forms an important part.

Alastair Brickell
Reply to  ptolemy2
January 21, 2018 5:16 am

January 21, 2018 at 3:25 am

I’m sure Dr.Ball is not a 6 day creationist. However, it matters not whether or not he is. The article is about ice cores and he has made some good points.

Importantly other commenters, notably Ferdinand Englebeen and others, have sensibly answered many of Dr. Ball’s points. Perhaps in time Dr. Ball will respond…I hope he does as he has lots to offer us.

However, the main point (for me at least) is that the discusion has been carried out on a sensible, calm and scientific basis. Very little of the name calling and abuse that you see on so many other sites…just point by point comments. This is what makes Anthony’s efforts creating and maintaining WUWT so very worthwhile. Long may it last and thanks to all involved as moderators and commenters.

Brett Keane
Reply to  Alastair Brickell
January 22, 2018 12:44 am

Alastair, true, and something Willis still has to learn….

January 21, 2018 4:20 am

I agree – ice core records, especially the air bubbles concentrations are very uncertain. This will be shown when the temperatures start decreasing in the next few decades.

Bernard Lodge
January 21, 2018 6:04 am

If the ice core results are reliable then the two main conclusions are:
1. Pre-industrial levels of CO2 were significantly lower than present levels and
2. Pre-industrial temperatures changed before CO2 concentrations changed
The first conclusion is an interesting historical fact. The second conclusion is much more important because it proves that CAGW is wrong. Changes in CO2 do not cause changes in temperature, it is the other way round!

By the way, the second conclusion from the ice cores is also confirmed with the modern day CO2 records from Mauna Loa which show that Atmospheric CO2 levels fluctuate significantly by season. This shows that seasonal temperature changes cause modern day CO2 concentration to change. Since I don’t think anyone would argue that changes in CO2 cause the seasons, then this also proves that changes in CO2 do not cause changes in temperature – it is the other way round!


A dependent variable cannot be an independent variable at the same time – unless there is a runaway effect. There obviously is no runaway effect otherwise we would not be here to talk about it.

Reply to  Bernard Lodge
January 21, 2018 6:38 am


A little careful: if the influence of two variables on each other is modest, they still can have that influence in both directions, without invoking a runaway reaction.
The natural effect of temperature on CO2 is about 16 ppmv/K for the dynamic equilibrium (“steady state”) between ocean surface and atmosphere.
The theoretical effect of a CO2 doubling (before any feedbacks) is about 1 K.
The current CO2 increase is far above the effect of temperature on CO2 levels, but even so, the effect of the extra CO2 on temperature may be largely masked by natural variability.

Alastair Brickell
Reply to  Ferdinand Engelbeen
January 21, 2018 12:55 pm

Ferdinand Engelbeen
January 21, 2018 at 6:38 am

Ferdinand you state upthread (@5:47) that “Humans add about 9 GtC (4.5 ppmv) per year of which about half remains in the atmosphere. That is over twice the natural variability and twice the observed increase in the atmosphere.”

How can we differentiate between mans’ fossil fuel CO2 and natural CO2 from plants? Are C13 ratios reliable and useful…what is the science behind the separation of human CO2 from natural CO2? Presumably C13 from old coal is the same as C13 from oil, biomass or forest burnoffs worldwide. It’s all plant (or algae) derived.

Reply to  Ferdinand Engelbeen
January 21, 2018 1:45 pm


The drop in δ13C in the past 165 years is enormous, compared to natural variability, much higher than even over a glacial-interglacial transition. Still that doesn’t prove that humans are to blame, as current plants have about the same δ13C level as ancient plants (in coal, lower in oil and much lower in natural gas).

But there is an alternative route by measuring the decrease of oxygen in the atmosphere (a hell of a job for an accuracy better than 1:200,000).
For each type of fuel, the oxygen use is known from burning efficiency. That shows that slighty less oxygen is used than calculated from fossil fuel use. That means that the biosphere as a whole (land+sea plants, bacteria, molds, insects, animals…) produce more O2 than consumed for plant decay, feed and food. Thus the biosphere is a net sink for CO2 and preferably 12CO2, leaving relative more 13CO2 in the atmosphere. Thus not the cause of the drop in δ13C. Neither are the oceans, as any increase of CO2 from the oceans would increase the δ13C level of the atmosphere…
See: http://www.bowdoin.edu/~mbattle/papers_posters_and_talks/BenderGBC2005.pdf

Alan Tomalty
Reply to  Ferdinand Engelbeen
January 21, 2018 10:58 pm

\Ferdinand Engelbeen/
\A little careful: if the influence of two variables on each other is modest, they still can have that influence in both directions, without invoking a runaway reaction./

The Mauna Loa data clearly show a statistically significant correlation between seasons and CO ppm levels in atmosphere assuming that CO2 is uniform (which I don’t accept) and also assuming that the data is uncontaminated ). However the warmists argue that we can believe the Mauna Loa figures. Okay then how can one argue that seasons have little or nothing to do with the constant seasonal variability that is quite evident from the basic graph of CO2? The data clearly shows this as Bernard Lodge has pointed out. Mankind may be responsible for 50% of the post industrial increase but it is impossible to argue that change in seasons is not driving the variability. The only factor that I can think of that would affect this seasonal variability beside temperature is change in the winds. If the warmists have to argue that possibility then they better provide wind data around Mauna Loa. If not; the basic Mauna Loa CO2 graph is proof positive that CO2 cannot drive temperature.

Brett Keane
Reply to  Ferdinand Engelbeen
January 22, 2018 1:01 am

Ferdinand, alas, even from you, the slide sideways to ‘CO2 effect on temp MAY be masked’. Feynmann noted that as a sign of a false claim. By such are the lies maintained…..

Reply to  Ferdinand Engelbeen
January 22, 2018 1:08 am

Alan Tomalty,

If you have several influences on the same variable, it is not that easy to separate the different causes and effects.

The natural variations have three main influences. mainly driven by temperature:

– Seasonal:
Countercurrent, + and – 60 GtC vegetation, – and + 50 GtC ocean surface.
Net effect: + and – 10 GtC (5 ppmv), mainly in the NH where seasonal growth and decay is dominant.
Increasing temperatures over the seasons give less CO2 in the atmosphere and reverse.

– Year by year:
Parallel: + 1 to 3 GtC in warm years (El Niñp), -3 GtC in cold years (Pinatubo) or +/- 1.5 ppmv around any trend. In general gets down to zero after 1-3 years.
Again (tropical) vegetation is dominant.
Increasing temperatures increase CO2 in the atmosphere at about 5 ppmv/K

– Very long term:
Based on ice cores, a rather fixed ratio of 8 ppmv CO2/K for polar temperatures. As there is the polar enhancement, the global impact over thousands of years is about 16 ppmv/K, mainly from the (deep) oceans. Not by coincidence app. the influence of temperature on the ocean surface – atmosphere dynamic equilibrium per Henry’s law.

In all natural cases, temperature drives CO2 levels, but the largest influence in fluxes is seasonal and that gives the opposite CO2 change as for the other influences: warmer gives less CO2 in the atmosphere…
There is little change in seasonal amplitude, a small increase over the past decades, thus the largest CO2 in/out fluxes remained about the same. Still the residual after a full seasonal cycle could be positive and contributing to the increase.
Based on the oxygen balance, that is not the case: except in El Niño years, the whole biosphere is a net sink for CO2, not a source.
Neither are the oceans, based on the higher δ13C level and Henry’s law.

Humans add 9 GtC/year, nature sinks 4.5 GtC/year, thus humans are fully responsible for (near) all of the CO2 increase in the past 60 years, except for a few ppmv caused by warmer ocean surface waters.

Any increase of CO2 increases the temperature at the surface, theoretically with about 1ºC for 2xCO2, independent of the source. In reality it may be much less (sceptics) or more (climate models), depending of the alleged negative or positive feedbacks. Anyway, that influence is independent of the influence of temperature on CO2 levels and does lead temperature in the case of human emissions.

Reply to  Ferdinand Engelbeen
January 22, 2018 1:11 am

Brett Keane,

Depends of the signal to noise ratio…

Take sea level changes from tide gauges: you need at least 25 years of data to show a statistical (!) change in sea level from the meters of level change by waves and tides…

Alastair Brickell
Reply to  Ferdinand Engelbeen
January 22, 2018 2:53 am

Ferdinand Engelbeen
January 21, 2018 at 1:45 pm

Thanks for the information but I’m still not quite sure just how this all works.

Why does excess O2 in the atmosphere mean that the “biosphere is a net sink for CO2 and preferably 12CO2”? Does oxidation of minerals come into play…eg. sulphur from volcanoes and other such processes like weathering? What is the link between the O2 and CO2 (or 12CO2)?

Was there a jump in 12C relative to 13C after WWII when the huge global increase in fossil fuels really got underway?

This is not my field so I need a simple explanation if possible! Are there intermediate steps/processes that I’m not grasping?

Reply to  Ferdinand Engelbeen
January 22, 2018 8:22 am


Step by step:

Burning fossil fuels uses oxygen, each type of fuel its own amount.
Sales of fossil fuels (taxes…) give you the global inventory of fossil fuel use.
Calculating fossil fuel use with their own oxygen consumption gives you how much oxygen is used globally.

Plants produce oxygen when they use CO2 from the atmosphere with photosynthesis.
Plant decay uses oxygen and produces CO2, as good as microbes, molds, insects and animals, all from the same plant base.
If there is more plant production than decay/feed/food, the oxygen level goes up and the CO2 level down. Or the opposite happens now and then (e.g. during an El Niño).

We can measure de decrease of oxygen over the years. That decrease is slightly less than what is calculated from the burning of fossil fuels. That means that plants produce more oxygen than the whole living biosphere uses. That also means that the CO2 balance of the biosphere is negative: more CO2 uptake than release.
As there is more CO2 uptake by plants than release by the rest of the biosphere and plants slightly prefer 12CO2 over 13CO2, that should increase the 13C/12C ratio in the atmosphere, but we see a firm decline, all caused by our use of fossil fuels:


Over the whole Holocene there was little variability (+/- 0.2 per mil) in the 13C/12C ratio in the atmosphere (ice cores), with the same variability in surface waters in the period 1350-1850 (in coralline sponges). From then on the levels dropped with over 1.6 per mil, due to the use of fossil fuels.
Ocean surface waters and atmosphere rapidly exchange CO2 with exchange rates of less than a year, thus any isotopic or CO2 level change in the atmosphere is near directly followed by a change in the ocean surface. The resolution of the sponges is 2-4 years, which is very fast and the δ13C level of the calcite in the sponges matches that of the surrounding waters quite exactly.

The increase of CO2 levels in the atmosphere and the drop in δ13C completely match each other, but only 1/3 of the original human emissions remains in the atmosphere, the rest is absorbed by and/or exchanged with other reservoirs (-deep- oceans and vegetation).

Other sinks and sources like rock weathering are minor sources/sinks and as far as I know didn’t substantially change over the past decades.