The foregoing uncertainty introduces a formidable obstacle to the determination of 

 absolute temperatures of the sea surface from aircraft using only infrared techniques. On 

 a particular occasion the uncertainity could be somewhat reduced either by estimating a 

 correction, or by directly measuring both the infrared signal from the portion of the sky 

 from which the reflected rays received by the radiometer originate and the sense and amount 

 of the heat flux across the thermal boundary layer. By sufficient effort, the correction might 

 be defined to within 0.5° K. 



What such uncertainty means in terms of displacement of the isotherms on a map of 

 the sea surface depends on the area considered. On the West Coast, where the horizontal 

 gradient of surface temperature may be as small as 1° K per degree of latitude, the displace- 

 ment of a given isotherm might, under severe conditions, be as great as 90 miles. On the 

 other hand, in a region where the horizontal gradient is very acute as along the western 

 boundary of the Gulf Stream, the displacement would be negligible. Since the correction is 

 altered by changes in the evaporation rate and the pattern of clouds, the apparent isotherms 

 as seen by radiometer must slither about the ocean surface like monstrous sea snakes. Any 

 representation of their configuration is in the nature of a snap-shot of an ever changing 

 scene. 



In any case, the horizontal temperature gradients are much more precisely defined 

 than is the absolute value of the particular isotherms. Thus, the spread or clumping of the 

 isotherms is correctly shown without regard to the temperature labels attached. For use in 

 marine biology and for many aspects of physical oceanography, the location of these transi- 

 tion boundaries may be far more significant than the absolute values of the temperature. For 

 example, it is well known that heavy fish populations occur in the neighborhood of these bound- 

 aries. This may be due to the 'snow fence' effect of the thermal barrier, or it may be the 

 result of food concentration in regions of horizontal convergence (Reference 3). Examples 

 of areas with high horizontal gradients that are highly productive are: the western edge cf the 

 Gulf Stream, the edge of the Kuroshio, the 'front' across the entrance to the Gulf of California, 

 and the region of the Antarctic convergence. 



It is therefore concluded that thermal maps of the sea surface made from infrared data 

 collected periodically from low-flying aircraft will be highly useful to marine biologists and 

 physical oceanographers, regardless of the difficulties of assigning absolute values to the 

 specific isotherms. As the state-of-the-art progresses, the real temperature will doubtless 

 be better and better defined. 



REFERENCES 



1. Ewing, G. C, and McAlister, E. D,, 1960. 'On the Thermal Boundary Layer of the Ocean', 

 Science, vol. 131, No. 3410, pp 1374 - 1376. 



2. McAlister, E. D., 1964, 'Infrared-Optical Techniques Applied to Oceanography: 1. 

 Measurement of Total Heat Flow from the Sea Surface', Journal of Applied Optics, vol. 

 3, No. 5, May 1964, pp 609 - 611. 



3. Hardy, C, 'The Open Sea', Collins, London, 1956, pp 335. 



-93- 



