Carbon cycle modelling and
the residence time of natural and anthropogenic atmospheric CO2 - 3

The atmospheric residence time (i.e. lifetime; turnover time) of CO2 has been quantified based on measurements of natural radiocarbon (carbon-14) levels in the atmosphere and the ocean surface; the changes in those levels caused by anthropogenic effects, like "bomb carbon-14" added to the atmosphere by nuclear explosions; and the "Suess Effect" caused by the addition of old carbon-14-free CO2 from combustion of fossil fuels; and the application of gas exchange theory to rates determined for the inert radioactive gas radon-222. The results from these measurements are shown in Table 2, mainly based on the compilation by Sundquist (1985), in addition to the solubility data of Murray (1992), and the carbon-13/carbon-12 mass-balance calculation of Segalstad (1992). Both of the last two recent methods happened to give a lifetime of 5.4 years based on completely different methods.

Table 2. Atmospheric residence time (i.e. lifetime, turnover time) of CO2, mainly based on the compilation by Sundquist (1985; for references in brackets).

Judged from the data of Table 2 there is apparently very little disagreement from early works to later works regardless of measurement method, that the atmospheric CO2 lifetime is quite short, near 5 years. This fact was also acknowledged early by IPCC's chairman Bolin (Bolin & Eriksson, 1959).

We should also note that a large number of the atmospheric CO2 lifetime measurements are based on anthropogenic additions of CO2 to the atmosphere by "bomb carbon-14". It is important for the understanding of the robustness of the ocean to deal with the anthropogenic extra CO2 that the measured lifetimes are within the same range as for natural carbon-14 before and after the nuclear bomb tests in the early nineteen-sixties. They are also coincident with lifetimes found when considering anthropogenic CO2 from Man's burning of fossil fuel, both from carbon-14 as well as for carbon-13/carbon-12 isotopes. The measured lifetimes in Table 2 therefore represent the real lifetime of atmospheric CO2 in dynamic contact with all its sources and sinks with "perturbations" included. Hence other "lifetimes" found by non-linear carbon cycle modelling are irrelevant.

The short atmospheric CO2 lifetime of 5 years means that CO2 quickly is being taken out of the atmospheric reservoir, and that approximately 135 giga-tonnes (about 18%) of the atmospheric CO2 pool is exchanged each year. This large and fast natural CO2 cycling flux is far more than the approximately 6 giga-tonnes of carbon in the anthropogenic fossil fuel CO2 now contributed annually to the atmosphere, creating so much political turmoil (Segalstad, 1992; 1996).

Supporters of the "Greenhouse Effect Global Warming" dogma have apparently not been satisfied with these facts based on numerous measurements and methods. They go on by saying that because we observe the atmospheric CO2 level increase, it must be caused by Man's burning of fossil fuel, and the "lifetime" of atmospheric CO2 must be 50-200 years (Houghton et al., 1990). Hence, they say, when we construct non-linear (non-proportional and non-chemical-equilibrium) non-steady-state systems for the fluxes between the ocean surface layer, the atmosphere, and the terrestrial system, the decay time of man-made carbon into the atmosphere must be much longer than the turn-over time (Rodhe & Björkström, 1979). Because if we now use a constructed evasion "buffer" factor (Section 5 and 6 above) of 10, the atmospheric CO2 "lifetime" will be 10 times the measured (real) lifetime of 5 years, namely 50 years or more (Rodhe & Björkström, 1979; Rodhe, 1992).

To rephrase; an apparent atmospheric CO2 level rise, assumed to be due to Man's burning of fossil fuel, is being treated with non-linear (non-proportional and non-chemical-equilibrium) non-steady-state modelling, giving theoretical far longer "lifetimes" than actually measured. When this is not explained to the readers, they are led astray to get the impression that the "artificial" un-real model "lifetimes" are real lifetimes.

Or as O'Neill et al. (1994) phrase it: "A growing array of timescales are being extracted from carbon cycle models and data and their relationships have not been clear." . . . "This discrepancy has not been adequately explained and is causing confusion in the literature concerned with the atmospheric "lifetime" of anthropogenic CO2" . . . "Considering the policy implications of such numbers, it is important that their meanings and relationships be fully clarified."

Rodhe & Björkström (1979) conclude their treatment of carbon cycle and CO2 lifetime modelling by: "Naturally, we do not claim that such very simplified models of the carbon cycle, which we have studied, contain the final answer to the very complex question of how nature will distribute the man-made CO2 emissions between the major reservoirs. That question should be studied with the aid of much more sophisticated models which take into account more of our knowledge about the physical and chemical processes involved."

10. The breakdown of the dogma - carbon isotopes

Suess (1955) estimated for 1953, based on the carbon-14 "Suess Effect" (dilution of the atmospheric CO2 with CO2 from burning of fossil fuel, void of carbon-14), "that the worldwide contamination of the Earth's atmosphere with artificial CO2 probably amounts to less than 1 percent". Revelle & Suess (1957) calculated on the basis of new carbon-14 data that the amount of atmospheric "CO2 derived from industrial fuel combustion" would be 1.73% for an atmospheric CO2 lifetime of 7 years, and 1.2% for a CO2 lifetime of 5 years.

This is in conflict with IPCC researchers, who assume that 21% of our present-day (as of December 1988) atmospheric CO2, the assumed rise in CO2 level since the industrial revolution, has been contributed from Man's burning of fossil fuel (Houghton et al., 1990).

This large contradiction between the carbon-14 measurements and the dogma, has worried many researchers. In order to make Suess' measurements fit the dogma, it would be necessary to mix the atmospheric fossil-fuel CO2 with CO2 from a different carbon reservoir five times larger than the atmosphere alone (Broecker et al., 1979). It was alternatively proposed that the carbon-14-labelled CO2 would act completely differently than the "ordinary" CO2: "However, the system's responses are not the same for the CO2 concentration and for isotopic ratios" (Oeschger & Siegenthaler, 1978). The explanation is given that the CO2 levels will be governed by the constructed evasion "buffer" correction factor, while on the other hand (strangely enough) the isotope ratios of the atoms in the very same CO2 molecules would be unaffected by the evasion "buffer" factor, and further:

"would be equal in both reservoirs [the atmosphere and the ocean's mixed layer] at equilibrium. This explains why the relative atmospheric CO2 increase is larger than the Suess effect" (Oeschger & Siegenthaler, 1978). This cannot be accepted, when all chemical and isotopic experiments indicate that equilibrium between CO2 and water is obtained within a few hours (see Section 5 above).

Ratios between the carbon-13 and carbon-12 stable isotopes are commonly expressed in permil by a so-called delta-13-C notation being the standard-normalized difference from the standard, multiplied by 1000. The international standard for stable carbon isotopes is the Pee Dee Belemnite (PDB) calcium carbonate.

CO2 from combustion of fossil fuel and from biospheric materials have delta-13-C values near -26 permil. "Natural" CO2 has delta-13-C values of -7 permil in equilibrium with CO2 dissolved in the hydrosphere and in marine calcium carbonate. Mixing these two atmospheric CO2 components: IPCC's 21% CO2 from fossil fuel burning + 79% "natural" CO2 should give a delta-13-C of the present atmospheric CO2 of approximately -11 permil, calculated by isotopic mass balance (Segalstad, 1992; 1996).

This atmospheric CO2 delta-13-C mixing value of -11 permil to be expected from IPCC's model is not found in actual measurements. Keeling et al. (1989) reported a measured atmospheric delta-13-C value of -7.489 permil in December 1978, decreasing to -7.807 permil in December 1988 (the significance of all their digits not justified). These values are close to the value of the natural atmospheric CO2 reservoir, far from the delta-13-C value of -11 permil expected from the IPCC model.

From the measured delta-13-C values in atmospheric CO2 we can by isotopic mass balance also calculate that the amount of fossil-fuel CO2 in the atmosphere is equal to or less than 4%, supporting the carbon-14 "Suess Effect" evidence. Hence the IPCC model is neither supported by radioactive nor stable carbon isotope evidence (Segalstad, 1992; 1993; 1996).

To explain this apparent contradiction versus the IPCC model, the observed delta-13-C value of atmospheric CO2 "must be affected by other heavier [i.e. with high delta-13-C values] carbon sources, such as is derived from the air-sea exchange process" (Inoue & Sugimura, 1985). One way to make this happen, would be if the isotopic exchange from air to sea were different from the isotopic exchange from sea to air; i.e. a gross non-equilibrium situation would be required. Siegenthaler & Münnich (1981) were able to construct such a simple theoretical kinetic, non-equilibrium model: "Diffusion of CO2 into the water, which is rate limiting for mean oceanic conditions, fractionates the carbon isotopes only little. 13-C/12-C fractionations are found to be -1.8 to -2.3 permil for atmosphere-to-ocean transfer, and -9.7 to -10.2 permil for ocean-to-atmosphere transfer."

Inoue & Sugimura (1985) attempted to verify these kinetic isotope fractionations experimentally at three temperatures: 288.2; 296.2; and 303.2 Kelvin, versus their equilibrium values of -8.78; -7.86; and -7.10 permil, respectively, all with uncertainty given as +/- 0.05 permil. Their reported air to sea fractionations at these temperatures were -2 +/- 3; -4 +/- 5; and -5 +/- 7 permil, respectively. Their sea to air fractionations were found to be -10 +/- 4; -13 +/- 6; and -12 +/- 7 permil, respectively. (Reported alpha fractionation factors and uncertainties have here been recalculated to alpha minus one, multiplied by 1000, to get comparable fractionation values). They conclude that the agreement is fairly good with the theoretically deduced values of Siegenthaler & Münnich (1981). Looking at the reported uncertainties, however, the experimental data cannot be grouped in three populations: their air-to-sea and sea-to-air data are not significantly different from their reported air/sea/air equilibrium value at the three different temperatures. Hence the experimental data cannot be used as evidence for the proposed theoretical difference in isotopic fractionation for air/sea versus sea/air CO2 transfer due to differences in kinetic isotope fractionation.

Siegenthaler & Oeschger (1987) touch in their carbon cycle modelling, with carbon isotopes included, on the possibility that the apparent atmospheric CO2 level increase is due to marine degassing instead of accumulation of anthropogenic CO2: "We will also discuss the sensitivity of the model results to uncertainties in the ice core data, to different model assumptions and to the (unlikely) possibility that the non-fossil CO2 was not of biospheric, but rather of marine origin." The word "unlikely" in parentheses is indeed their wording. Their modelling shows ambiguously that: "as expected, the results are similar to those for the fossil-only input". But their modelling shows a discrepancy with the ice core CO2 data, in addition to: "it is somewhat surprising that observations and model agree for 13-C but not for 14-C; this can, however, not be discussed here any further". In their abstract, however, they conclude on the contrary: "Calculated 13-C and 14-C time histories agree well with the observed changes."

The carbon cycle modelling of Siegenthaler & Oeschger (1987) run into several problems making their models fit all the data, leading them to write: "One possibility is that the assumptions underlying our results are not fully correct, i.e., that either the Siple ice core data deviate from the true atmospheric concentration history or that the carbon cycle models used do not yield the correct fluxes. If we dismiss these possibilities, then other carbon sinks than the ocean seem to exist." For the lack of validity of the Siple ice core, see Section 4 above.

Based on this kind of modelling, IPCC states as part of their "evidence that the contemporary carbon dioxide increase is anthropogenic" (their Section 1.2.5; Houghton, 1990): "Third, the observed isotopic trends of 13-C and 14-C agree qualitatively with those expected due to the CO2 emissions from fossil fuels and the biosphere, and they are quantitatively consistent with the results from carbon cycle modelling." Such a correspondence is, however, not evident to the present author.

Segalstad (1992; 1993; 1996) concluded from 13-C/12-C isotope mass balance calculations, in accordance with the 14-C data, that at least 96% of the current atmospheric CO2 is isotopically indistinguishable from non-fossil-fuel sources, i.e. natural marine and juvenile sources from the Earth's interior. Hence, for the atmospheric CO2 budget, marine equilibration and degassing, and juvenile degassing from e.g. volcanic sources, must be much more important; and the sum of burning of fossil-fuel and biogenic releases (4%) much less important, than assumed (21% of atmospheric CO2) by the authors of the IPCC model (Houghton et al., 1990).

The apparent annual atmospheric CO2 level increase, postulated to be anthropogenic, would constitute only some 0.2% of the total annual amount of CO2 exchanged naturally between the atmosphere and the ocean plus other natural sources and sinks (Section 9 above). It is more probable that such a small ripple in the annual natural flow of CO2 is caused by natural fluctuations of geophysical processes. We have no database for disproving this judgment (Trabalka, 1985). Like Brewer (1983) says it: "Nature has vast resources with which to fool us . . .".

Segalstad's mass balance calculations show that IPCC's atmospheric CO2 lifetime of 50-200 years will make the atmosphere too light (50% of its current CO2 mass) to fit its measured 13-C/12-C ratio. This proves why IPCC's wrong model creates its artificial 50% "missing sink" (Segalstad, 1996).

11. Conclusion

The atmospheric CO2 level is ultimately determined by geologic processes. The carbon on the Earth's surface has come from CO2 degassing of the Earth's interior, which has released about half of its estimated CO2 contents throughout Earth's history during the 4,500 million years up to now (Holland, 1984). Important geologic processes are volcanism and erosion, releasing carbon from the lithosphere and the Earth's interior to the atmosphere - ocean - biosphere system. These processes are counteracted by sedimentation of carbonate and organic carbon in the hydrosphere (mainly the ocean). The balance between these two main processes determines the CO2 level in the atmosphere (e.g., Kramer, 1965; McDuff & Morel, 1980; Walker & Drever, 1988; Holmén, 1992). "Thus, while seawater alkalinity is directly controlled by the formation of calcium carbonate as its major sedimentary sink, it is also controlled indirectly by carbonate metamorphism which buffers the CO2 content of the atmosphere" (McDuff & Morel, 1980).

In addition there is a short-term carbon cycle dominated by an exchange of CO2 between the atmosphere and biosphere through photosynthesis, respiration, and putrefaction (decay), and similarly between aqueous CO2 (including its products of hydrolysis and protolysis) and marine organic matter (Walker & Drever, 1988).

Analogously to the transfer of anthropogenic CO2 to the atmosphere, it seems appropriate to cite Walker (1994): "Consider, now some perturbation of the system - for example, the doomsday perturbation that suddenly stops photosynthesis. In 20 years or so, all the carbon in the biota reservoir will be released to the atmosphere, leading initially to a large increase in the amount of carbon dioxide in the atmosphere. But in no time at all, in terms of human generations, that extra carbon dioxide will work its way down into the very deep sea reservoir where the addition of 2 x 10¹⁷ moles to the 30 x 10¹⁷ moles already there will have little effect. The system will not end up with a lot of extra carbon dioxide in the atmosphere, even if photosynthesis stops completely. The figure shows also the fossil fuel rate, which is smaller than the rate of photosynthesis."

It is nature's coupling between the temporary, short-lived atmospheric reservoir, with 0.5 x 10¹⁷ moles CO2, and the relatively enormous oceanic reservoir, with 30 x 10¹⁷ moles of dissolved (and hydrolyzed and protolyzed) CO2 in contact with calcium carbonate, that determines the amount of CO2 in the atmosphere. This coupling is in turn coupled to the much larger lithospheric reservoir. The rates and fluxes of the latter coupling control the amount of carbon in the surface reservoir of the Earth. All kinds of measurements show that the real residence time of atmospheric CO2 is about 5 years.

Chemical and isotope equilibrium considerations and the short CO2 residence time (lifetime) can fully explain the carbon cycle of the Earth. The conclusion of such reasoning is that any atmospheric CO2 level rise beyond 4% cannot be explained by accumulation of CO2 from Man's burning of fossil fuel. An apparent CO2 rise can only come from a much larger, but natural, carbon reservoir with much higher delta-13-C than the fossil fuel pool, namely from the ocean, and/or the lithosphere, and/or the Earth's interior. CO2 degassing from the oceans instead of IPCC's anthropogenic accumulation is indeed made probable by the measurements of a larger CO2 increase in Atlantic surface waters than in the contemporaneous atmosphere (Takahashi, 1961; 1979). Kondratyev (1988) argues that: "The fact is that the atmospheric CO2 content may be controlled by the climate" and not the opposite.

Trabalka (1985) concluded: "The available data on past fluctuations in atmospheric CO2 and climate suggest that our current carbon cycle models, which emphasize human perturbations, may be missing natural feedback components involving both terrestrial and marine systems, perhaps even climate-induced "mode switches" in ocean circulation patterns, which could be very important in understanding changes in both climate and the carbon cycle over the next century."

Such conclusions will not make the large "doomsday" headings in the news media, will not make the politicians implement extra taxes or legislations, will not make expensive conferences organized by the United Nations or other international bodies, will not make environmental organizations preach about the wickedness of Man, and will not bring any research support money from governments or research foundations.

Carbon cycle modelling and CO2 - 2 - 3 - 4

http://folk.uio.no/tomvs/esef/ESEF3VO2.htm