Temperature differences and their variations with depth from 2.6 Nansen 

 casts and BT's taken simultaneously in the upper layers in September i960 

 at station CHARLIE are listed in Table 1. The range of surface tempera- 

 ture differences between the Nansen casts and BT's was 2.1° F (0.2 to 

 2.3°); the range of differences below the thermocline was 2.35 F (0.15° 

 to 2. 50°); and the range of variation between the surface difference and 

 the difference below the thermocline was 0.9° F (from 0° to 0.9°). 



TABLE 1 



TEMPERATURE DIFFERENCES AND VARIATIONS BETWEEN SIMULTANEOUS 

 NANSEN AND BT CASTS AT THE SURFACE AND BELOW THE THERMOCLINE 



SURFACE 



BELOW THERMOCLINE 



RANGE 



Temperature 

 difference (°F) Percent 



Temperature 

 difference (°F) Percent 



Variation 

 (°F) 



Percent 



0-0.5 



15 



0-0.5 



15 







27 



>o.5 



85 



>o.5 



85 



0-0.5 



62 



>1.0 



38 



>1.0 



50 



0.5-0.9 



11 



>1.5 



27 



>1.5 



35 







>2.0 



15 



>2.0 



15 







Seven different bathythermographs were used in this series of 26 

 observations. The results may be considered fairly representative of 

 most of the BT data. The error of temperature differences measured 

 between the sea surface and any depth in the lower part of the thermo- 

 cline or below the thermocline is generally smaller than 0.5 F. 



Depth can be read on a ^50-foot BT trace with accuracy of approxi- 

 mately 5 feet, yet little is known about the reliability of the existing 

 BT data with respect to depth. There are no references similar to bucket 

 and injection temperatures for checking depth. 



There is a case in the Hydrographic Office data where the USS SAN 

 PABLO and USS REHOBOTH operated only a few miles from each other. The 3^ 

 BT's taken on the REHOBOTH during one day placed the mean layer depth at 

 157 feet with an oscillation of 25 feet about the mean. Approximately 

 the same number of BT's taken by the SAN PABLO during the same time inter- 

 val, however, shows the mean layer depth at 128 feet with about the same 

 degree of oscillation. This indicates a difference in mean layer depth 

 of 29 feet. Apparently one bathythermograph had lost calibration. 



There is no way of determining the number of similarly inconsistent 

 BT's in the existing data. Some evidence could be collected to compare 

 the mean thermocline depth computed from the last 5 or 6 BT's of a station 



