of thermocline; changes of temperature between 75 and 125 meters due to the 

 intrusion of cold or wann water; and minimum temperature differences below 

 125 meters. The last figure indicates differences in structure between the 

 first and last BT observations, a time interval of 26 days, 3 hours, and a 

 space interval of 47 nautical miles (84 km) . The temperature in the iso- 

 themnal layer decreased from 21.0°C to 19 .8°C during the period of observa- 

 tion. Although the amplitude of the thermocline depth oscillations was 

 almost the same on the first day as on the last, the mean depth of the 

 thermocline incr-eased from 84 to 102 feet between September 2nd and 28th. 



The decrease of temperature in the isothermal layer, and the increase in 

 mean depth of the thermocline are in agreement with the expected seasonal 

 changes at this location during the month of September. 



In order to convert the T-D variations into a continuous spectrum, the 

 depths of occurrence of individual isotherms were read from the analog 

 thermistor chain record. Figures 4 and 5 are plots of the selected iso- 

 therms for two periods of 30 hours, based on the 6-minute thermistor chain 

 data record. The relation of the selected isotherms to the T-D structure is 

 illustrated in Figure 6 . 



In these figures, no attempt has been made to indicate the existence of 

 the temperature inversions shown in Figure 3, primarily because their full 

 extent was not visible on the 300 foot (91 m) thermistor chain records. The 

 appearance and disappearance of the temperature inversions, however, does 

 appear, from the deeper BT observations, to be related to the low frequency 

 internal waves which are clearly seen in Figures 4 and 5 at shallower depths, 



These two sections of the total record were selected to illustrate that 

 there were appreciable differences in the height of the low frequency 

 internal waves at these two phases of the moon. The maximum height of the 

 low frequency internal wave at the top of the thermocline, as shown by the 

 oscillations of the 20°, 17° and 15.5° isotherms, occurred at the time of 

 the full moon (Fig 5) . The height of the wave at the bottom of the thermo- 

 cline was at near maximum at the three-quarter moon phase (Fig 4 and 

 Table 1). The maximum height of the wave at the bottom of the thermocline, 

 110 ft., however, occurred on September 16th, one day prior to the time of ■ 

 the full moon. The height of the wave at the top of the thermocline at this 

 period was only 48 feet . 



The mean depth of the 17° isotherm was the same during the two periods 

 but the standard deviation was larger during the full moon. The mean depth 

 of the 12.5° isotherm was 35 feet deeper at full moon but the standard 

 deviation was smaller at this time (Table l) . 



Table 2 summarizes statistics of the depth variability of the 5 selected 

 isotherms for the entire series . The standard deviation is least for the 

 17° isotherm--15.45 ft., and largest for the 12.5° isotherm--18 .38 ft. The 

 range of observed depths increases from 66 ft. for the 20° isotherm to 148ft. 

 for the 12.5* isotherm. Tables 1 and 2 indicate that variability of the 



410 



