ClimateGate news

Sunday, March 25, 2007

Beck's 138 year-long record of atmospheric CO2

A report by Ernst-Georg Beck, "180 YEARS OF ATMOSPHERIC CO2 GAS ANALYSIS BY CHEMICAL METHODS" published by Energy & Environment (Vol 18 No. 2 2007) compares 90,000 accurate chemical analyses of CO2 levels in the air recorded from 1812 through 1961. This record shows a remarkedly different trend compared to the literature of UN's International Panel on Climate Change (IPCC).

Here's a rather lengthy, but interesting excerpt:


More than 90,000 accurate chemical analyses of CO2 in air since 1812 are summarised. The historic chemical data reveal that changes in CO2 track changes in temperature, and therefore climate in contrast to the simple, monotonically increasing CO2 trend depicted in the post-1990 literature on climate-change. Since 1812, the CO2 concentration in northern hemispheric air has fluctuated exhibiting three high level maxima around 1825, 1857 and 1942 the latter showing more than 400 ppm.

Between 1857 and 1958, the Pettenkofer process was the standard analytical method for determining atmospheric carbon dioxide levels, and usually achieved an accuracy better than 3%. These determinations were made by several scientists of Nobel Prize level distinction. Following Callendar (1938), modern climatologists have generally ignored the historic determinations of CO2, despite the techniques being standard text book procedures in several different disciplines. Chemical methods were discredited as unreliable choosing only few which fit the assumption of a climate CO2 connection.


The causes, development and future projection of climate change are summarized in the reports of the Intergovernmental Panel on Climate Change (IPCC), a United Nations body that is responsible for advising governments. The four consecutive Assessment Reports of the IPCC - issued in 1992, 1995, 2001 and 2007 – follow closely the views of three influential scientists, Arrhenius, Callendar and Keeling on the importance of CO2 as a control on climate change. Quote from Keeling (1978, p. 1 [1]).
The idea that CO2 from fossil fuel burning might accumulate in air and cause a warming of the lower atmosphere was speculated upon as early as the latter half of the nineteenth century (Arrhenius, 1903). At that time the use of fossil fuel was too slight to expect a rise in atmospheric CO2 to be detectable. The idea was again convincingly expressed by Callendar (1938, 1940) but still without solid evidence of a rise in CO2.
Following this line of argument, the IPCC’s Third Assessment Report (IPCC, 2001, chapter 3.1 [2]) contained the further explanation which makes it entirely explicit that direct measurements can only be relied on post 1957 and prior direct measurements can be disregarded in favour of indirect measurements made of air trapped in ice:
The concentration of CO2 in the atmosphere has risen from close to 280 parts per million (ppm) in 1800, at first slowly and then progressively faster to a value of 367 ppm in 1999, echoing the increasing pace of global agricultural and industrial development. This is known from numerous, well-replicated measurements of the composition of air bubbles trapped in Antarctic ice. Atmospheric CO2 concentration have been measured directly with high precision since 1957; these measurements agree with ice-core measurements, and show a continuation of the increasing trend up to the present.
In 1958 C.D. Keeling, University of California, San Diego, USA, introduced a new technique for the accurate measurement of atmospheric CO2. Keeling used cryogenic condensation of air samples followed by NDIR spectroscopic analysis against a reference gas, using manometric calibration. Subsequently, this technique was adopted as an analytical standard for CO2 determination throughout the world, including by the World Meteorological Association (WMO) [3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13].

CO2 measuring stations are distributed across the globe. Most, however, are located in coastal or island areas in order to obtain air without contamination from vegetation, organisms and industrial activity, i.e. to establish the so-called background level of CO2. In considering such measurements, account should be taken of the established fact that land-derived air flowing seawards looses about 10 ppm of its carbon dioxide to dissolution in the oceans, and even more in colder waters (Henrys Law).


A major issue regarding the IPCC approach to linking climate and CO2 is the assumption that prior to the industrial revolution the level of atmospheric CO2 was in an equilibrium state of about 280 ppm, around which little or no variation occurred. This presumption of constancy and equilibrium is based upon a critical review of the older literature on atmospheric CO2 content by Callendar and Keeling. (See Table 1).

Between 1800 and 1961, more than 380 technical papers that were published on air gas analysis contained data on atmospheric CO2 concentrations. Callendar [16, 20, 24] Keeling and the IPCC did not provide a thorough evaluation of these papers and the standard chemical methods that they deployed. Rather, they discredited these techniques and data, and rejected most as faulty or highly inaccurate [20, 22, 23, 25, 26, 27].

Though they acknowledge the concept of an ‘unpolluted background level’ for CO2, these authors only examined about 10% of the available literature, asserting from that that only 1% of all previous data could be viewed as accurate (Müntz [28, 29, 30], Reiset [31], Buch [32]).


During my own review of the literature, I observed that the evaluation of Reiset’s and Müntz’s work by Callendar and Keeling was erroneous. This made me investigate carefully the criteria that were used by these and other authors to accept or to reject such historical data.

The data accepted by Callendar and Keeling had to be sufficiently low to be consistent with the greenhouse hypothesis of climate change controlled by rising CO2 emissions from fossil fuel burning. Callendar rejected nearly all data before 1870 because of “relatively crude instrumentation” and reported only twelve suitable data sets in 20th century as known to him [20] out of 99 made available by Stepanova 1952 [18]. The intent of these authors was to identify CO2 determinations that were made using pure unpolluted air, in order to assess the true background level of CO2. Callendar set out the criteria that he used to judge whether older determinations were “allowable” in his 1958 paper [20] which presents only data that fell within 10% of a longer yearly average estimated for the region, and also rejected all measurements, however accurate, that were “measurements intended for special purposes, such as biological, soil air, atmospheric pollution”.

Next I cite the conclusion of the analysis of 19th centuries CO2 data by Keeling back in 1986 (From/Keeling 1986, pp. 101–103 [23]):
“Our original goal was to find, if possible, a seasonal cycle in the nineteenth century atmospheric CO2 data in agreement with modern observations by applying the air mass criteria of Callendar (1940a) to screen out contaminated data. This goal we have demonstrated to be unachievable.

We find, after screening out suspicious data on the basis of air mass, that none of the five data sets of Callendar show the seasonal cycle which Callendar found in combination.

Brown and Escombe (1905b) investigated atmospheric carbon dioxide only as a slide line to botanical studies. They provide minimal information on methodology and weather conditions. A few of their data seem abnormally low. Their sampling was sporadic over a four year period at a site poorly chosen to study CO2, albeit convenient to their botanical laboratory. Their results are of interest mainly because they used an apparatus similar to Reiset’s which had been carefully tested by an independent method.”

“In conclusion, the measurements of atmospheric carbon dioxide carried out by Reiset (1882) from 1872 to 1880 on the coast of northern France appear to be valid. They indicate a mean annual concentration, with respect to dry air, of 292.4 ±1.2 ppm. Comparisons with other possibly valid contemporary data suggest that these data are not biased by more than 10 ppm. It is thus unlikely that the CO2 concentration was less than 282 ppm in the late nineteenth century, and was probably close to 292 ppm.”
There was no verification or falsification of results and methods used by other authors, especially those published in the 20th century (e.g. Lundegardh [35, 36], Duerst [37], Kreutz[38], Misra [39], Scholander [40]), with exception of Buch 1935 [32], lying on the “fuel line” (Callendar 1958 [20]).

According to Callendar, Keeling and the IPCC, CO2 variations to be observed in air were due diurnal, and seasonal cycles, or to glacial/ interglacial fluctuations. Natural concentrations are assumed to have been in equilibrium until mankind disturbed the natural situation. In this way, any long term observations that might display decadal to centennial natural variations in atmospheric CO2 are ruled out a priori by Callendar and Keeling.

As I discuss further below, these criticisms by Callendar and Keeling, and the selective way in which they discarded previous data, are not able to be justified. Their most egregious error was perhaps the dismissal of all data which showed variations from their presupposed average. That said, it is of course the case that some of the older data has to be viewed as less reliable for technical, analytical reasons, as also indicated below.


In this paper, I have assembled a 138 year-long record of yearly atmospheric CO2 levels, extracted from more then 180 technical papers published between 1812 and 1961. The latter year marked the end of the era of classical chemical analysis.

The compilation of data was selective. Nearly all of the air sample measurements that I used were originally obtained from rural areas or the periphery of towns, under comparable conditions of a height of approx. 2 m above ground at a site distant from potential industrial or military contamination. Evaluation of the chemical methods used reveals systematically high accuracy, with a maximum 3% error reducing to 1% for the data of Henrik Lundegardh (1920–26), a pioneer of plant physiology and ecology [34, 35, 36].



During the late 20th century, the hypothesis that the ongoing rise of CO2 concentration in the atmosphere is a result of fossil fuel burning became the dominant paradigm. To establish this paradigm, and increasingly since then, historical measurements indicating fluctuating CO2 levels between 300 and more than 400 ppmv have been neglected.

A re-evaluation has been undertaken of the historical literature on atmospheric CO2 levels since the introduction of reliable chemical measuring techniques in the early to middle 19th century. More than 90,000 individual determinations of CO2 levels are reported between 1812 and 1961. The great majority of these determinations were made by skilled investigators using well established laboratory analytical techniques. Data from 138 sources and locations have been combined to produce a yearly average atmospheric CO2 curve for the northern hemisphere.

The historical data that I have considered to be reliable can, of course, be challenged on the grounds that they represent local measurements only, and are therefore not representative on a global scale. Strong evidence that this is not the case, and that the composite historical CO2 curve is globally meaningful, comes from the correspondence between the curve and other global phenomena, including both sunspot cycles and the moon phases, the latter presented here probably first time in literature and the average global temperature statistic. Furthermore, that the historical data are reliable in themselves is supported by the credible seasonal, monthly and daily variations that they display, the pattern of which corresponds with modern measurements. It is indeed surprising that the quality and accuracy of these historic CO2 measurements has escaped the attention of other researchers.

How to interpret the monthly variation of CO2 (see Fig. 5, 7, 9 and modern measurements e.g. Mauna Loa), which indicates a coincidence with the lunar phases, is another question to be dealt within a paper in preparation.

Modern greenhouse hypothesis is based on the work of G.S. Callendar and C.D. Keeling, following S. Arrhenius, as latterly popularized by the IPCC. Review of available literature raise the question if these authors have systematically discarded a large number of valid technical papers and older atmospheric CO2 determinations because they did not fit their hypothesis? Obviously they use only a few carefully selected values from the older literature, invariably choosing results that are consistent with the hypothesis of an induced rise of CO2 in air caused by the burning of fossil fuel. Evidence for lacking evaluation of methods results from the finding that as accurate selected results show systematic errors in the order of at least 20 ppm [28, 29, 30, 31, 57, 73]. Most authors and sources have summarised the historical CO2 determinations by chemical methods incorrectly and promulgated the unjustifiable view that historical methods of analysis were unreliable and produced poor quality results [2, 20, 22, 23, 24, 25, 26, 27, 65, 74, 95].
Here is a link to a review of Beck's report.

See also:
Greenie Watch: Real History of Carbon Dioxide Levels
An Englishman's Castle: Ernst-Georg Beck's paper 180 Years accurate CO2 Gas analysis of Air by Chemical Methods


JR said...

Very interesting. Have you seen any attempt to rebut this?

EliRabett said...

Trivial. Look at the rapid changes in CO2 concentrations in Beck's graph. As many have pointed out that is impossible. Too much CO2 moving in and out of the atmosphere too fast to be believed.

JoeG said...

EliRabett - "Trivial" to dismiss because it isn't possible!??

Fortunately most scientists find that "impossibility" as a starting point for research.

There is a definite anomoly in temperatures around 1940. The CO2 measurements taken then correlate to that.

The obvious question is WHAT is going on here. A great opootunity to investigate the relationship of CO2 to temperature.