For centuries it was commonly believed that the Earth's climate is, and always has been, stable. The idea that the climate, on the contrary, has always changed has its origins with Ignaz Venetz who, in 1821, laid the foundations of our modern knowledge of the ice ages.
The acceptance of glacial advance and retreat demands acceptance of a mechanism which will cause such events. That mechanism is climate change. The fact that our planet's climate has always changed is now a foundation - and a very sure foundation - of modern climate science.
The Swedish scientist Svante Arrhenius was the first to suggest that fossil fuel combustion may eventually affect global average temperature. He proposed a relation between atmospheric CO2 concentrations and average global temperature. He found that the average surface temperature of the earth is about 15oC due to the infrared absorption capacity of water vapor and carbon dioxide - the greenhouse effect. Arrhenius built his theory on a foundation laid by others, most notably a paper that his colleague Arvid G. Högbom published in 1894.
In the Physical Society of Stockholm there have been occasionally very lively discussions on the probable causes of the ice age; and these discussions have, in my opinion, led to the conclusion that there exists as yet no satisfactory hypothesis that could explain how the climatic conditions for an ice age could be realized in so short a time as that which has elapsed from the days of the glacial epoch. The common view hitherto has been that the Earth has cooled in the lapse of time; and if one did not know that the reverse has been the case, one would certainly assert that this cooling must go on continuously. Conversations with my friend and colleague. Professor Högbom, together with the discussions above referred to, led me to make a preliminary estimate of the probable effect of a variation of the atmospheric carbonic acid on the temperature of the Earth. As this estimation led to the belief that one might in this way probably find an explanation for temperature variations of 5o- I0o C., I worked out the calculation more in detail, and lay it now before the public and the critics.
Svante Arrhenius, “On the Influence of Carbonic Acid in the Air upon the Temperature of the Ground”, 1895
The global carbon cycle as described by Högbom consists of many separate and independent Processess. Högbom theorised that a stable level of atmospheric CO2 was unlikely to persist for a geologically long period. He described how CO2 is sequestered in sedimentary formations in the form of limestone and other carbonates, how it is bound in plants and is absorbed in the oceans. He then considered the decomposition of plants, the weathering of rocks and volcanic activity as natural processes tending to restore CO2 to the atmosphere. He also suggested that the combustion of carbonaceous meteorites in the atmosphere could be a minor source of atmospheric CO2. Högbom concluded that the amount of atmospheric CO2 was likely to have varied considerably over time, as the result of these different processes.
It was from Högbom's theory that large variations in the atmospheric concentration of CO2 were quite likely in different geological periods that Svante Arrhenius hypothesised that the onset of ice ages and interglacials results from natural variations in the level of atmospheric CO2.
Högbom's Om sannolikheten för sekulära förändringar i atmosfärens kolsyrehalt seems never to have been translated into English. The translation presented below was produced with the help of Google Translate, bab.la and a very old book of Swedish grammar. Given the changes in Swedish word usage and spelling since 1894 it was necessary to make some assumptions as to meaning, based on context. Also, the term 'carbonic acid' and like terms have been replaced with 'CO2', for brevity. The paragraphs have been numbered for ease of reference. It is to be hoped that this translation is accurate enough to be useful to all who have an interest in climate science and its history.
Note that the idea of crustal shrinkage, as mentioned in footnote 1, was replaced by plate tectonics theory.
Translation commences below.
On the probability of global changes in the level of atmospheric CO2 (1894)
By Arvid G. Högbom
A translation into modern English.
By Patrick Lockerby
1 In view of the weighty role which atmospheric CO2 plays as an agent in the metabolism of both inorganic and organic natural processes, it is likely that this matter will be of a certain interest to the chemist: whether a steady state obtains In regard to atmospheric CO2, or whether the processes by which CO2 is supplied to the atmosphere, and the processes by which it is extracted from thence, are sufficiently independent of each other, such that appreciable changes would be assumed to take place over the course of long periods of time. Its principal role is nonetheless an issue for geologists; I have come to concern myself somewhat with some questions in connection with geological sequestration in limestone. It may seem strange that the question is on the global variability in the level of atmospheric CO2, although knowledge about it in many respects would be of the greatest importance for the perception of Earth's evolutionary history, not yet the subject of thorough studies or reflections. But it is clear on the other hand, that because of the many variables, and the scale of this difficult problem and the subtle factors, which are accordingly to be taken into account, the question can hardly be fully answered. But one will be able to get somewhat further than the dubious , or in many cases obviously false conclusions about this matter often found in textbooks and popular works.
2 Even if it is not possible to obtain accurate quantitative expressions of the reactions by which CO2 is produced and consumed, yet certain factors are discovered, of which one should be able to form an approximately correct idea, and from which some illuminating conclusions can be drawn regarding the core issue.
3 It appears that the first matter of importance is to compare the existing atmospheric amounts of carbon dioxide with the converted amounts. If the former relative to these is negligible, then the probability for variations is obviously different than in the opposite case.
4 The question as to the present level of atmospheric CO2 can be calculated with sufficient accuracy. The numerous determinations made at different seasons, at different parts of the earth's surface and at different heights above sea level show the CO2 ratios within suitably narrow limits, and should not be far from the actual average level, about the same as added to 0,03 Vol% for the whole of that part of the atmosphere in which the ordinary air circulation takes place. As this circulation, however, reaches several thousand meters in altitude, and thus encounters the greatest part of the atmospheric mass, so it cannot be said that the measurement would be considerably modified, even if within the very topmost layers of air another level of CO2 content might exist, especially if the variation is not particularly great.
5 The atmosphere has an average concentration of CO2 0.03 Vol%, so this number represents 0.045 percent by weight, 0.342, etc. partial and 0.466 gr. C02 over every cm 2 of earth surface. If we imagine the atmosphere's entire CO2 component as bonded with calcium, we would obtain a 3.7mm thick limestone layer over the entire surface of the earth, and, reduced to carbon, the atmospheric CO2 would give an approximately 1mm. thick layer. The amount of carbon which is found bound in the living organic world, can certainly not be estimated with nearly the same degree of reliability, but obviously, the number which would be likely to express that amount will be of the same order, so that the atmospheric level of CO2 can neither be regarded as very large or very small in comparison with that bound within organisms. In relation to the rapid metabolism of organic nature, the remaining atmospheric CO2 available is thus not so great that it cannot change, as could take place by climatic or other grounds, considering that this turnover rate and amount, could produce disorders in the stable equilibrium.
6 The following calculation is also quite illustrative for estimating the relationship between the atmospheric CO2 fraction and the amount of CO2 which is generated by human activity. Earth's current coal production is, in round numbers 500 million tonnes per year or 1 ton per km2 of earth's surface. Converted to CO2 it represents an amount about one thousandth of the total of atmospheric CO2. It corresponds to a limestone layer of 0.003 mm across the globe, or 3 mm over Sweden, or, expressed in cubic capacity, 1.5 km3 of limestone. This CO2 component which is supplied to the atmosphere chiefly by modern industry, can be considered to fully compensate for the quantity which, as by weathering and silicate rock decomposition, is used up in the formation of limestone and other carbonate rocks. One has felt entitled to estimate the additional quantities of CO2, which in dissolved form would be carried under the sea in the course of one year, namely nearly 3 km3 per year , by determining the level of dissolved CO2, especially as carbonates, in the estimated water volume in a number of rivers in different countries and even climates by calculating their drainage area relative to the earth's entire land surface. Since there are also found rivers whose drainage areas take up silicate, transporting very insignificant quantities of calcium carbonate relative to the rivers that take it up in limestone areas, so one can draw, even by other substantiated grounds, the conclusion that only a small proportion of these 3 km3 carbonates are formed immediately by decomposition of silicate, or in other words, that only a small part of this quantity of calcium carbonate can be formed in the course of one year by weathering processes. If also, because of the incomplete and uncertain data on which it is estimated, the proposed number were incorrect by 100% or even more, however it makes a comparison of fairly great interest since it shows that the most important of all the processes by which CO2 at all times left the atmosphere, namely, the chemical weathering of silicate rock, is quantitatively comparable to, or of the same order of magnitude as, a process of the opposite kind, which is provoked by the industrial developments of our time and must be considered be of a transitory nature.
7 Compared with that bound in limestones (and other carbonate rocks), the amount of CO2 in the air is vanishingly small. In view of the thickness of sedimentary formations and the significant part that limestone or carbonate rocks form of them, it is not impossible that the total amount of carbonate would represent a layer round the globe of hundreds of meters. Assuming for example the thickness to be 100 meters one calculates about 25,000 times more carbon dioxide, which is linked to the calcium in the sedimentary formations, than the free carbon dioxide in the atmosphere. Every CO2 molecule in this limestone mass, however, has been in and passed through the atmosphere in the past. (The assumption of 100 meters, however, may well be incorrect by 100% or more, but by using that number the probability of a large over-estimate will be less than the probability of an under-estimate.) Even leaving aside all other factors that could affect the atmospheric CO2 level, this fairly small number lends probability to the assumption that it would, in older geological times, have varied between limits which only slightly depart from the contemporary concentration. As the weathering processes consumed amounts of CO2 which were many thousands of times larger than that now available in the atmosphere, and then, besides, these processes because of different geographical, climatological and other conditions probably proceeded with unequal intensity for various times, apparently the probability of significant variability in atmospheric CO2 would be quite large, even if account is taken of the compensation processes which developed, according to what is shown in the following, when for one reason or another supply or consumption of CO2 tended to increasingly upset stable equilibrium. You often see the opinion expressed that the level of atmospheric CO2 during ancient geological periods would have been much greater than in the present, and that the reduction would be from this cause: that the carbon dioxide in the form of coal and carbonates had been removed from the atmosphere and deposited in the earth's crust. In many cases this hypothetical reduction is ascribed solely to coal formation, while the much more important carbonate formation is completely overlooked. That the former process does not play a particularly appreciable role thereof will be apparent given that even at a relatively high assessment of the Earth's available coal supply the quantities are negligible in comparison with existing quanties of limestones. Looking at it this way: if one of the CO2 reduction processes caused constant losses in the atmosphere, though spread over the course of time, enormous amounts of CO2 would have been fixed in carbonate rocks but if one goes into a close examination of the facts there are processes by which in all ages CO2 has been added to the atmosphere. Of these processes, one can well conclude that it is likely that considerable variation has taken place, but not the contrary, that the change occurred in a specific direction.
8 Carbon dioxide can be considered to be supplied to the atmosphere chiefly through the following processes:
1) volcanic emissions, together with the following geological phenomena,
2) combustion within the higher air layers of carbonaceous meteors,
3) organic substances burning and decaying,
4) decomposition of carbonate rocks,
5) liberation of the carbon dioxide absorbed in sea-water as a result of temperature increase or reduction in atmospheric CO2 and partial pressure,
6) liberation of mechanically entrapped carbon dioxide in rocks when they decompose by weathering etc.
The consumption of atmospheric carbon dioxide is chiefly limited to the following processes: the formation of carbonate rock, siliceous rock weathering, global plant assimilation and the absorption of carbon dioxide in sea-water. The processes designated 4) and 6) play a quite secondary quantitative role to the first category, with 6) counting less than 4). Likewise the sea water's CO2 content, the level of which is essentially determined by atmospheric carbonation, can be left aside, however, with the remark that the absorption and release of CO2 by the sea must be a regulatory influence on atmospheric CO2, the global variations of which thus become slower and fewer with an increase or decrease in supply (with respect to consumption), with only a small part of its amount altering the level of atmospheric CO2, inasmuch as the sea is capable of absorbing far more CO2 than that currently existing in the atmosphere. Combustion and decay processes could also, with some restrictions, have been omitted. The natural organic carbon cycle occurs for the most part with such a great speed, that the degree of variability therein does not to a particularly noticeable degree have any impact on the level of atmospheric CO2. If, nevertheless, as the case seems to have been within certain geological periods, organic substances escape the normal ongoing processes of combustion or decomposition in great amounts and are stored in the sedimentary formations as coal etc, this will obviously be a not insignificant factor, as these stored products, through combustion by contemporary industry is, as regards quantity, remarkable according to what is often demonstrated.
9 It is even likely that organic life, as well as sea-water, is a regulatory influence on atmospheric CO2 in so far as the total amount of the global organically bound carbon stock grows in the same direction as atmospheric CO2 levels. Meteor falls are a CO2 source of absolutely indeterminable significance. Given that no more than a significant percentage are composed of carbon or carbon substances, their frequency and probability does, however, permit a presumption that they could supply great amounts of CO2 to the atmosphere. This CO2 source is different from the others listed in two notable respects: it increases the Earth's total carbon stock, and it supplies CO2 to the atmosphere's very topmost layers, which lie outside the ordinary air circulation. Volcanic eruptions, during the course of which the CO2 is incorporated which has its origin in the Earth's slow annealing of rock over time - from liquid to solid state, probably of all the recounted processes this matter may be of the greatest importance. The molten lava in the earth contains, among other absorbed gases, large amounts of CO2 which escape through seepage, or when the lava is pushed up to the surface and the pressure is reduced. Since the immense quantities of CO2 - corresponding to several atmospheres of pressure - which are now bound up in limestones in the Earth's crust, cannot have left even a tiny fraction simultaneously existing in the atmosphere since organic life has appeared on Earth, unless CO2 reduction by weathering and carbonate reservoir formation has been offset by successive supplies. It follows that volcanic emissions can be assumed to have first provided CO2 to the atmosphere.
10 This CO2 source has likewise, to all appearances, not flowed steadily and uniformly. Just as individual volcanoes show variability in periods of relative rest and intensive activity, so apparently the globe as a whole has in certain geological epochs shown a more violent and general volcanic activity or, on the other hand, has been distinguished by a reduction in manifestations of volcanic forces1. It is therefore quite probable that the atmospheric CO2 level has undergone thereby approximately simultaneous variability, or at least before now, this has been a factor that could significantly influence the CO2 level.
11 If one takes an overview of the CO2 formation and the CO2 consuming reactions discussed above, they will be found to be not so obviously in a relationship with and dependent upon each other that any probability exists for a long term capacity for equilibrium in regard to atmospheric CO2. Rather, if the reactions in question are considered individually and in relation to their contemporary atmospheric CO2 quantities, might they not be so disposed that a range of considerable variations ought to take place within, geologically speaking, short times. Whether or not such a notion is well founded will best be seen from the following considerations.
12 Suppose for example that a sudden supply of CO2 equal in amount to that formerly existing in the atmosphere took place in one way or another. This would by no means cause the atmospheric CO2 to be doubled. Since the greater part of the supply would be counterbalanced by the result of the increased partial pressure causing increased CO2 absorption in sea-water, in reality atmospheric levels of CO2 would become only slightly increased, and in addition through the same mechanism increased CO2 binding in the mineral realm would be induced, and possibly even increased assimilation in the plant world, so that the atmosphere further tends to return to the previous CO2 level. In a similar tendency for the maintenance of an equilibrium, capacity for consumption would be caused to decrease if the supply were suddenly reduced. If in the former case, most of the excess CO2 was stored in the sea, the absorbed CO2 in the sea would in such a case serve as a reservoir, which for a time would induce a replacement for most of the decreased supply elsewhere. One can here in quite a literal sense speak of an "economy of nature" 2.
13 Notwithstanding the dependency, in which parts of the CO2 forming and CO2 consuming processes thus stand to each other, and by which variations in atmospheric CO2 become to some degree balanced or spread over a long time, however, the likelihood that the ratio varies within limits little distinguished from the current CO2 level is not particularly great. This increase or decrease in supply taking place only over geological periods must, even if the increase or decrease is not great, cause substantial changes in atmospheric CO2, and there is not any conceivable obstacle to this being, in a geological epoch, several times greater or on the other hand much less than now. The possibility of such changes should not be overlooked when seeking explanations for petrographic characteristics, which distinguish certain geological formations or strata. As a detailed conclusion on the relevant issues may not be of any special interest to this magazine's readership, I shall confine myself to only a few short hints regarding a couple of the most important issues, which at the same time are partly of a chemical nature.
14 The quantity of carbonates (principally calcium carbonate), which all of earth's rivers supply to the sea during the course of one year is assessed, according to the previously stated grounds, at almost 3 km3. It would seem readily apparent, when one knows the sea-basins' overall cubic capacity and the approximate solubility in sea-water of these carbonates, that within the course of some thousands of years the sea would become saturated in regard to said carbonates so that afterwards an immediate crystallization of these would take place, corresponding to the further added carbonate amount. The fact that this saturation state is not entered into depends on the sea organisms constant secretion of calcium carbonate. There would seem as regards the carbonate content of sea water to be a tendency to bring about an approximately stable equilibrium, analogous to what was said before concerning atmospheric carbonation: an increased supply leads without doubt to a more abundant carbonate secretion by organisms so that the increase to a final saturation level is, so to speak, opposed.
15 However, if during a geological era the atmosphere had, for example, a few times higher CO2 than during others, so also ought the rivers during this time, under the assumption of things being roughly equal, partly because of intense weathering. Secondly and principally because a much more abundant direct dissolving of crustal carbonate rocks would have brought to the sea vastly greater amounts of carbonates, so that saturation could occur and limestone or dolomite be directly crystallized. Accumulation conditions in terms of the sedimentary formations in limestones are certainly in many respects shrouded in mystery, but perhaps it may be that for some special cases or formations it could be demonstrated that such a direct crystallizing has taken place, while the larger main body of Earth's limestones are most likely to be interpreted as having been formed by limestone organisms and their detritus.
1) What has just been mentioned does not contradict the ultimate cause of these phenomena: the Earth by its slow cooling. Uniformity In regard to that cooling requires not that the manifestations of it in the form of volcanic phenomena shall take place uniformly. Only when the contraction produced by the temperature reduction reached a certain point did the crust react by folding and dislocations, which opened the way to the volcanic eruptions and exhalations. It needed afterwards some time before these processes developed once again.
2) Fully reliable determinations of CO2 absorbed in sea-water are certainly too few in number that one could make a comparison between the sea's and the atmosphere's absolute CO2 levels, but investigations made so far seem, however, to suggest that the sea's free and half-bound CO2 is very considerable in its amount compared with atmospheric CO2.
Suggestions for an improved translation are welcome.