1. The equations of Joseph and Sender (1958), which were em- 

 ployed in computing the dispersion of radioactive wastes due to turbu- 

 lent diffusion, are based on a statistical concept which provides a 

 smoothed space and time distribution. A time record of the concentra- 

 tion at any given point, or the spatial record of the concentration at 

 any given time resulting from a single actual point source, would differ 

 in a random manner from the smoothed distribution predicted by the 

 equations of Joseph and Sender. There would thus occur, from any 

 single discharge, periods of time during which the concentration over 

 small areas would be higher than that predicted. Since there would 

 also occur corresponding times and locations with concentrations less 

 than predicted, and since our concern is, at least in part, with the in- 

 tegrated effect of a number of such releases over time and space, this 

 departure of the actual distribution from the predicted distribution does 

 not introduce serious error in the final computations. 



2. In the evaluation of the open sea environment, a 100 meter 

 thick stirred layer was assumed. In some ocean areas this may be too 

 large. In the case of a 10 meter stirred layer, the estimates of the 

 time, tpp^- , required for the concentration to be reduced to environ- 

 mental ppc levels would be increased by a factor equal toVTO, or approx- 

 imately 3.2; the corresponding value of N, t he a llowable number of dis- 

 charges per month, would be decreased by vTo5 , or approximately by a 

 factor of 32. Such a small value of the layer depth for the open sea 

 would be very unusual; any real over estimation of this factor is most 

 probably compensated for by the neglect of transport of any activity to 

 the deeper water. 



3. In the computations for coastal waters, the possible concen- 

 tration of activity on the bottom, due to uptake by suspended silts and 

 subsequent settling, was not included. While this would be a conserva- 

 tive factor for those organisms which spend most of their time swim- 

 ming or drifting in the water, it may well be quite detrimental to bot- 

 tom living forms, particularly detritus and filter feeders. Data are 



not available to evaluate adequately the significance of this phenomenon. 

 It is primarily for this reason that our final recommendations relative 

 to inshore waters are more conservative than might otherwise be war- 

 ranted from a strict application of our numerical results. 



MONITORING AND RECORD KEEPING 



It is essential that a systematic monitoring program be initiated 

 as soon as possible to determine the consequences of the release of 

 radioactive wastes from nuclear-powered vessels, both civilian and 

 military*. This program is required in order to protect public health 

 and property, to modify regulations in the light of new knowledge, and 

 to prepare for intelligent action if and when nuclear disasters occur in 

 inshore environments. 



*A monitoring program has also been recommended by the Committee on Oceanography of 

 the National Academy of Sciences - National Research Council (in its report entitled 

 "Oceanography 1960-1970"). 



48 



