T H E Q U E S T I O N O F I N C R E A S E O F A T M O S P H E R I C CO, 23 It might be tempting to assume that the account for an atmospheric reservoir of more effects found in the samples investigated and than 1.5 times the CO, in the atmosphere. their individual variations are due to local Nevertheless, it is interesting, in order to contamination of air masses by industrial CO, demonstrate how the uncertainty in the size and that the world-wide decrease in the C14 of the atmospheric carbon reservoir affects our activity of wood is practically zero. This, how- results, to introduce “effective” reservoirs A* ever, implies a too fast exchange rate and is and S*, so that A* > A, because of exchange inconsistent with the lower limit of t (atm) with soil carbon, and S* < S, because of given above. By coincidence, it so ha pens incomplete mixing in the oceans. A* and S* that taking the average of the Y* values isted, are defined in such a way that Eq. (I) becomes 1.73 %, an exchange time z (atm) of 7 years is obtained as an upper limit, identical with the lower limit obtained previously from the C14 age of marine surface material. An ex- For Eq. (5) : change time of 7 years, however, makes it necessary to assume unexpectedly short mixing times for the ocean.’ By reconsidering the relative size of the marine and atmospheric carbon reservoirs, and is to be substituted. Then the relationships of the three unknown the assumptions necessary for treating the combined reservoirs as a closed system, we quantities k,, S* and A* according to the two may approximate more closely to the condi- equations (I a) and (5 a) are shown graphically tions prevailing in nature. First, it seems in figure I . With an average age of surface sea carbon possible that some of the organic matter and carbonate present in soils may have to be added of 400 years, i.e., k, = 11400,one finds, even to the carbon that constitutes the atmospheric with an improbably large A* equal to 3 A, carbon reservoir. Practically nothing is known that S* is only 10 to 30 % smaller than S, so about the C14 age of soils and it may be that that the resulting mixing times of the oceans some of the soil carbon, partly through cannot be larger than a few IOO years. The overturn time for the marine carbon bacterial action, is in more rapid isotopic exchange with atmospheric CO, than the data through rock weathering and precipitation can on plant assimilation and biological oxidation be estimated to be of the order of IOO,OOO years indicate. The total amount of carbon in soils (table 3). During an average lifetime of sea may be equal to, or larger b as much as a carbon of 400 years an amount corresponding factor of two than that in t e CO, of the to 0.3 % of the marine reservoir will be added atmosphere. Rapid exchange with such carbon to the oceans through rock weathering in the would decrease the change in C14 activity form of C14 free carbon and be precipitated resulting from industrial fuel combustion by a after isotopic equilibration. On the average, factor equal to the ratio of the total “effective” this w d cause the C14 age of sea carbon to atmospheric carbon reservoir to the atmos- appear about 35 years older. Locally, and pheric CO,. However, the overturn time of especially along shores, the effect may well be the atmospheric CO, through the terrestrial considerably greater, so that the ages measured biological cycle and the soil is probably at so far may, on the average, be too great by least several decades and it therefore seems as much as IOO years. In figure I, solutions improbable that a mechanism exists that might are shown for I/k,=300 years, indicating a value of S * / S of about for acceptable values of A*.1 In a separate paper in this issue, H. Craig evaluates A possibility that the addition of C14 free the exchange time by considering present data on the CO, to the atmosphere from volcanic eru G 1 4 production rate by cosmic rays, and obtains a value of 7 1 3 years. He further finds that a more detailed tions in the 19th century may obscure t e P il model of ocean mixing gives the same value. An earlier estimate of 20 to 50 years (SUESS1953) was based on C-14 measurements of woods grown in an area of industrial contamination. Tellus IX (1957). 1 r For a discussion of the physical significance of S*/S in terms of mixing times in the ocean see H.Craig’s paper in this issue. 4