This is the desired expression for true ocean surface temperature. Obviously, if ^ 

 and t were both unity, the expression reduces to T = T . In practice, t can be taken as 0.98 



and t can be obtained from the curves as indicated previously. T , the air temperature, is 



measured and these three terms substituted into the above equation to "correct" the indicated 

 temperature, thus giving (approximately) the true ocean surface temperature. 



The foregoing has dealt with the problem of accurately measuring the temperature of 

 the surface of the ocean. What is meant by "surface"? According to Ewing and McAlister^, 

 the absorption in the region from 6/x to 20m by water is so high that 98% of the absorption 

 occurs in the first 0.1 mm. In other words the radiation measured by the Infrared Therm- 

 ometer is received from the first three or four one-thousandth"s of an inch of ocean surface. 

 Thus, we are measuring the very top of the ocean and the temperature of this micro-surface 

 is different than the temperature of the sub-sui-face water. Ewir^ and McAlister performed 

 some measurements in 1959 off Scripps' Pier at the University of California, and found that 

 the micro-surface was approximately 0.6°C cooler than water 15 centimeters below the 

 surface. They reported that the conditions of wind and humidity were not conducive to vig- 

 orous evaporation; although they did not record the temperature and relative humidity at the 

 time of their measurements, the evaporative cooling of the surface could presumably be 

 considerablj^ higher under certain conditions. They then submerged a pump which welled 

 the 15 cm. deep water up to the surface to "rupture" the surface. The radiation tempera- 

 tures then rose approximately to the same temperatures as those measured by a thermo- 

 meter submersed at the pump intake. When the pump was shut off, the radiation tempera- 

 ture of the surface dropped about 0.6°C in about 12 seconds. 



The ocean, in ice-free latitudes, is heated to a considerable depth by short wave- 

 length solar radiation. The heat balance is maintained largely by evaporative cooling and 

 long wavelength re-radiation from the micro -surf ace. More work needs to be done with 

 regard to the difference in temperature between the micro-surface and the sub-surface 

 under different conditions of weather. Until further information is at hand, the best we can 

 do is to carefully note temperature, relative humidity, and wind conditions at the time of 

 all of our measurements, in addition to sub-surface temperatures as obtained by lightships, 

 if possible. 



PRACTICAL HARDWARE 



This section deals with some of the practical problems which may be encountered 

 in usir^ the hardware. Generally, this consists of those problems of getting one of several 

 models of the Infrared Thermometer which have been produced, to operate in the airborne 

 environment with its noise, wind, vibration, temperature fluctuations, and lack of correct 

 electrical power. There is also the problem of operatir^ a chart recorder under the same 

 circumstances. The early Barnes Infrared Thermometer (the IT-1) utilized batteries to 

 bias the detector. These batteries exhibited an operating life on the order of 60 to 100 

 hours. The Model IT-2, which succeeded the IT-1, was fundamentally identical in principle 

 except for a more readable output meter and an electronic bias supply which eliminated 



-33- 



