4. RADIOISOTOPES AND LARGE-SCALE OCEANIC MIXING 



W. Broecker 



1. Introduction 



Considerable information regarding the patterns and rates of large-scale 

 circulation can be obtained from a knowledge of the distribution of various 

 radioactive isotopes in the sea. Both naturally produced radioisotopes with 

 half-lives of the same order of magnitude as the time constants for oceanic mixing 

 processes and man-made radioactive isotopes, produced in sufficient quantities 

 by nuclear tests, are useful in this type of study. The natural isotope approach 

 is most valuable in the study of deep water masses, where residence times of 

 more than 100 years are involved, and the artificial isotopes for problems related 

 to near-surface water-masses, which undergo much more rapid mixing. 



As the circulation of the ocean is exceedingly complex, radioisotope data 

 alone are not adequate to establish mixing rates and patterns. It is only when 

 the data are combined with evidence from classical oceanographic research 

 that any progress can be made. The most useful method of combining these 

 two quite different types of information is to set up models of oceanic mixing 

 based on classical observations and theory. The isotope data are used first to 

 reject certain of these models and then to establish absolute values for various 

 rate parameters in the remaining models. Obviously if the distribution of more 

 than one isotope is studied, more stringent limits can be set up allowing more 

 sophisticated models to be treated. To date the isotope approach is in its in- 

 fancy ; the necessary data are still sparse and the models adopted greatly over- 

 simplified. Nevertheless, several important conclusions can be drawn from the 

 available work. 



2. Useful Isotopes 



To be suitable for large-scale oceanic mixing studies, an isotope should have 

 the following ideal characteristics : ( 1 ) it should be present in measurable 

 quantities in all parts of the ocean, (2) differences in concentration well outside 

 the limits of measurement error should exist, (3) the mode and rate of injection 

 of the isotope into the system must be known both as a function of time and of 

 space, (4) the isotope should move with the water acting as an infinitely 

 soluble salt, and (5) contributions of natural and artificial production of the 

 isotope should be distinguishable. 



Only four naturally occurring isotopes, 3 H, 14 C, 32 Si and 226 Ra, have been 

 given serious consideration in connection with circulation studies. Natural 

 tritium, which potentially should be an excellent tracer, suffers from two very 

 serious difficulties: (1) because of its low production rate (<2 T atoms cm -2 

 sec -1 ) and short half-life (~12 years), the abundance of tritium in the deep 

 water masses is more than an order of magnitude below the present limits of 

 experimental detectability, and (2) bomb tests and nuclear plant leaks have 

 added an amount of artificial tritium probably exceeding that previously 



[MS received October, 1960] 88 



