information, we can rewrite the depth prediction equation to contain a 

 tidal correction term in addition to the wind effect term 



Dw,T,t = 37.5 - 2.6 Wt-15, + 4 (T - Tt_3) Equation (2) ■ 



where 37.5 is the zero wind depth (in feet) adjusted to account for the 

 constant discrepancy present in equation (l) and (T - T^-_3) is the tidal 

 oscillation about the mean 3 hours prior to time t. 



A comparison of the diurnal depth cycle, as predicted by equation (2), 

 with the measured diurnal depth cycle shows very close agreement (Fig 8). 

 The tidal effect on the predicted diurnal depth tends to decrease the 

 high amplitude diurnal cycle that would be expected if only wind effects 

 were considered. 



In addition to the results obtainable from the essentially quanti- 

 tative survey, one fact of thermocline motion has been brought out by 

 direct inspection of the 4-month record of thermocline depth, i.e., 

 tidal oscillations about the mean thermocline depth decrease in 

 amplitude as the general level of the thermocline approaches either 

 the upper or lower water boundaries. This has also been observed to 

 occur for shorter period internal waves (LaFond, 1961). 



DISCUSSION OF ANALYSIS 



To determine more clearly influences of wind and tide on thermocline 

 motion, and to establish whether or not significant features of the 

 motion occur that are not directly related to either wind or tide, a 

 power spectra analysis was carried out on 33 days of unbroken hourly 

 data. The spectra that were calculated are listed below and presented 

 in Figures 9 and 10: 



Hourly measured thermocline depth data, (Dm t) 

 Hourly predicted depths using equation (l) (Dw,t) 

 Hourly values of Dm^t - Dw,t " ^t 

 Hourly measured surface tide, T^. 



A study of the spectra reveals a long-term trend of the thermocline 

 motion that is partially, but not entirely, a wind effect. This is 



126 



