TIME: 148.00 HOURS. TAU X: 0.00 TAU Y: 

 DEPTH (CM): 2072 CLOCK HOUR: 19 



0.37 WIND DIR AND SPD: 550 MEXED LAYER 



U 



DIR 



TRANSPORT (M) 

 E N 



TEMP 



DENSITY 







3.87 



-1.26 



4.07 



108.0 



17051 



5 



3.87 



-1.26 



4.07 



108.0 



17049 



10 



3.87 



-1.26 



4.07 



108.0 



17048 



15 



3.87 



-1.26 



4.07 



108.0 



16739 



20 



3.87 



-1.26 



4.07 



108.0 



14277 



25 



3.70 



-3.31 



4.96 



131.9 



11836 



30 



3.70 



-3.31 



4.96 



131.9 



8781 



35 



3.70 



-3.31 



4.96 



131.9 



5621 



40 



3.70 



-3.31 



4.96 



131.9 



1428 



45 



0.00 



0.00 



0.00 



0.0 







50 



0.00 



0.00 



0.00 



0.0 







1505 



1465 



1397 



1219 



263 



-857 



-1041 



-250 



188 











8.35 

 8.35 

 8.35 

 8.35 

 8.35 

 7.85 

 7.89 

 7.89 

 7.89 

 6.21 

 6.17 



1.02460 

 1.02460 

 1.02460 

 1.02460 

 1.02460 

 1.02473 

 1.02473 

 1.02473 

 1.02473 

 1.02534 

 1.02547 



Figure 3.2 {Continued ) Model output after a period of variable winds. 



RESULTS 



In testing the model, I used a two phase proce- 

 dure, due to the unavailability of combined cur- 

 rent and thermocline data. I first tested the 

 mixed layer deepening portion, then the current 

 portion. The combined result should be valid 

 because of superposition. The basic assumptions 

 of the two sections are comparable. 



1. Variations in the Thermal Structure 

 I tested the equations dealing with the varia- 

 tions in the thermal structure against data taken 

 at Ocean Station Papa, as presented by Denman 

 and Miyake (1973). I ran my model for both the 

 two day storm period of 21-23 June 1970 and the 

 longer record of 13-20 June. Because of the gen- 

 eralized form of the radiative flux which I used, 

 the agreement between the data and my model 

 lacks during some days, but overall the fit is good. 

 Figure 4.1 shows the values obtained for mixed 

 layer deepening over the Gaussian-shaped storm 

 of 21-23 June, and Figure 4.2 shows the fit of the 

 sea surface temperature prediction for the same 

 period. 



Overall the fit between the curves is very good. 

 The temperature curve shows a +0.1 degree C 

 bias towards the end of the record. This is due to 

 the fact that the actual value of the radiative flux 

 was about one half of the average value I used in 



my computations. This caused the predicted 

 temperature to be higher than the actual tem- 

 perature due to the higher flow of heat into the 

 model. 



The coefficients I required for this fit were 

 close to those found by Denman and Miyake 

 (1973). The value I found for the amount of wind 

 energy available for mixing as .0014, close to the 

 Denman and Miyake value of .0012, and close to 

 the value they calculated from Kato and Phillips 

 (1969) of .0015. The value for the extinction coef- 

 ficient was .002 cm"\ 



The other area of comparison was to the long 

 data record of 13-20 June. (Figures 4.3, 4.4) I 

 compared both the sea surface temperature and 

 bathythermal profile during the entire period. 

 Again, the overall fit is good. The areas of great- 

 est departure on 16 and 17 June are due to the 

 low actual values of the radiative input for those 

 dates, causing the predicted temperature to be- 

 come higher than the actual temperature for 

 those dates. Late on 14 June (local), the marked 

 difference in temperature is due to an abnor- 

 mally high radiative flux late in the daylight 

 hours of the 14th, which my generalized radia- 

 tion format did not take into account. For an 

 operational use, however, this generalized radia- 

 tion boundary condition is necessary, because 

 such data is not commonly available at sea. 



