Clark and Yarnall 



time distribution of the short period internal wave signatures, broken 

 down into the indicated ranges of period. These were included in the 

 hope of observing a relationship between the occurrence of these signa- 

 tures and the environmental parameters. No obvious relationships are 

 visible. 



It will be noted in Figure 24 that the wind components have 

 been time shifted relative to the phase and temperature curves. This 

 is to remove the effect of the delay between the time that a north 

 wind is established and the time the effects are felt in the H3 prop- 

 agation path--a delay of about two days is assumed. We should expect 

 that the wind must act on the water for a certain period of time be- 

 fore an effective volume of surface transport can be established. 

 Several days of persistent north winds are apparently required. 

 Figure 24 shows that the first step function increase in H3 phase, 

 occurring on October 29, is concurrent with the first (time shifted) 

 period of persistent north winds. Previous to that time north winds 

 were of short duration; winds were generally from the east or south- 

 east, and the step effect is not observed. 



The correspondence of events in H3 phase, N-S wind, and 

 thermistor T5 is most evident in the period October through December. 

 Figure 25 provides for a direct comparison of these three parameters 

 during this time. Subsequent to the event of October 29, the details 

 of the visual correlation may be followed both in H3 phase and in the 

 thermistor string data that is available. The correlation is apparent- 

 ly weaker in much of December and January. At the end of the test the 

 net change in H3 phase, referred to the reference level of October 6, 

 is a positive 8 cycles. According to the calibration of Figure 4, 

 this suggests a spatially averaged net increase in sound speed 

 (temperature) in the H3 propagation path from October to January. 



A single event provides the most convincing direct evidence 

 now available for cross-stream transport of surface water into the 

 H3 path. The longest period of persistent north winds in available 

 data occurred from 11 through 26 November. We will examine a detail 

 of the thermistor chain data (Fig. 26) for the latter part of this 

 period. (The detail starts at the position of the arrow in Figure 25). 

 At the beginning of the detail the medium is thermally stratified, as 

 indicated by the thermistor T5 curve. Subsequently, and we will as- 

 sume under the action of the north wind, a complete mixed layer forms. 

 This step is essentially completed by 23 November. Following this 

 step the entire mixed layer at the thermistor chain rises in temper- 

 ature about 0.6°C. This is the third stage in a process which re- 

 quires an external source of relatively warm water. Cross- stream 

 transport is the probable source of this warmer water. 



Turning now to the data of hydrophone H43 , we may suspect, 

 from the contrasting phase behavior at H43 and H3 , that the long 

 period signatures at H43 are predominately the result of events oc- 

 curring in the main stream of the Florida Current beyond H3 . 



331 



