316 



REVELLE AND SITESS 



[OHAP. 7 



the carbon dioxide produced by fossil-fuel combustion remained in the atmos- 

 phere. He came to this conclusion by com])aring COq analyses of air made in 

 the 19th century with those made more recently. Detailed statistical evalua- 

 tions have been made by Slocum (1955) and Bray (1959) with widely differing 

 conclusions as to the statistical significance of the available data. 



As was shown in a previous paper by the authors (Revelle and Suess, 1957), 

 one can conclude from radiocarbon measurements that the rate of exchange 

 and uptake of CO2 by the oceans must be of the order of 10 years. Arnold and 

 Anderson (1957) arrived independently at the same conclusion. Other in- 

 vestigators (Bolin and Eriksson, 1959; Bolin, 1960; Broecker, Tucek and 

 Olson, 1959; Craig, 1957, 1958; Rafter and Fergusson, 1957) attempted to 

 derive a more precise figure for this exchange rate, or more accurately, for the 

 residence time of CO2 in the atmosphere, by evaluation of the empirical measure- 

 ments by more rigorous calculations. All the considerations were based on the 

 measurements of two empirical quantities: (1) the apparent i^C age of ocean 



ATMOSPHERE 



i-^lf- 



No 



MIXED LAYER 

 1,2 N, 



STRATOSPHERE 

 TROPOSPHERE 



i^lf- 



OCEAN 

 58 Na 



BIOSPHERE 

 0.5 Aio 



-If— 



HUMUS 

 1.5 A/o 



Fig. 1. Carbon exchange reservoirs. N^ denotes the number of C atoms in the atmosphere. 

 For each exchange reservoir the number of C atoms in terms of Na is shown. 



water and (2) the effect of industrial fuel combustion on the specific activity 

 of atmospheric CO 2. However, our knowledge of both these quantities seems 

 now less firm than it was thought to be a few years ago. A more detailed 

 mathematical treatment cannot, therefore, appreciably improve the precision 

 of the results. 



Fig. 1 shows schematically the various "exchange reservoirs" of carbon and 

 their relative size, taking CO2 of the atmosphere as unity. The ocean is divided 

 by the thermocline into a mixed layer and deep water. Craig (1958) discusses a 

 "chain model" and a "cyclic model". The first considers transport of CO2 into 

 the deep sea through the mixed layer [(1) in Fig. 1] ; the second also takes into 

 account direct exchange with the deep water in polar waters [(2) in Fig. 1] 

 where no thermocline exists. The third alternative, viz. no downward mixing 

 through the thermocline and exchange exclusively occurring in polar waters 

 was used by Broecker, Gerard, Ewing and Heezen (1960). Craig (1958) obtains 

 a value for the residence time in the atmosphere of 7 + 3 years. Broecker, 

 Gerard, Ewing and Heezen (1960) obtain a similar value and also discuss non- 



