SECT. 2] GASES 319 



and the rate of its uptake into the ocean. Since 1954, large amounts of "arti- 

 ficial" ^-^C have been introduced into the atmosphere, primarily into the 

 stratosphere. At present, mixing of tropospheric and stratospheric air has not 

 been completed and at sea-level the i^C concentration in the atmospheric CO 2 

 is still rising. In the spring of 1960, the ^^C in the troposphere had risen to more 

 than 30° Q above its pre-bomb 1953 value. The increase due to bomb testing 

 was first published by Rafter and Fergusson (1957) of the New Zealand labora- 

 tory, and has since been observed by most i^c laboratories. The level of i^C 

 in the Southern Hemisphere lags about eighteen months behind that of the 

 Northern Hemisphere. If and when there is a cessation of atmospheric testing, 

 the artificial excess ^^C will, over the following years, equilibrate with that of 

 the oceans, so that the atmospheric ^^C activity will again decrease steadily 

 down to near normal levels. After some time required for the mixing of the 

 stratospheric and tropospheric air-masses, the rate of decrease will correspond 

 to the mean residence time of CO 2 in the atmosphere. At the same time the 

 14C in the surface water of the oceans will rise markedly. Broecker and Olson 

 (1950. 1960) estimated the expected changes in ^^C concentration under 

 the assumption that testing had ceased in 1960. According to these authors the 

 atmospheric i^C activity would have reached a maximum around 1964, and 

 average surface ocean water could have been expected to reach a maximum 

 14C activity, of about 15% above the present, around 1975. Because of the un- 

 certainties in the basis of these estimates, a considerably different course of 

 events would not be surprising. 



An accurate knowledge of the residence time of CO2 in the atmosphere and 

 rate of absorption into the sea must be combined with other factors in order 

 to predict the effect of fossil-fuel combustion upon the CO2 content of the 

 atmosphere in future times with desired precision. In particular one has to 

 consider the peculiar buffer mechanism of sea-water, which causes an increase 

 in the partial CO 2 pressure of more than an order of magnitude higher than the 

 increase in the total CO 2 concentration when CO 2 is added and the alkalinity 

 remains constant (Revelle and Suess, 1957 ; Kanwisher, 1960). This affects not 

 only the equilibrium distribution of CO2 between atmosphere and ocean but 

 also the rate of CO 2 uptake in particular in the case of relatively slow mixing 

 through the thermocline. This circumstance increases the amount of CO 2 from 

 industrial fuel combustion in the atmosphere. 



In any case, radiocarbon measurements have shown that a large fraction of 

 the CO 2 released by industrial coal combustion ^in any case more than 50%) 

 has been taken up by the ocean. To arrive at a more quantitative figure and 

 make more accurate predictions as to the future increase of CO 2 in the atmos- 

 phere seems at present impossible. In order to make precise predictions of this 

 secular increase, one will probably have to go back to the original way sug- 

 gested by Callendar, which is simply to compare precise analytical data taken 

 over many years. Because of their poor accuracy, the data from analyses of 

 CO 2 in air made in the 19th century cannot be used for this purpose. A very 

 precise series of measurements started during the IGY by Keeling (1960) will 



