Sea computations. Thus, in an attempt to overcome our present problems, the sound 

 speed at the top of the second layer was increased to approach that in the North Sea 

 model. A sound speed of 1600 m/s allowed convergence for modes of propagation with a 

 consistent set of phase velocities. 



Computed propagation losses are shown in Figure 4.13 for the winter profile. As 

 evidenced by the anomalously high losses, we still have problems with the computational 

 model. The failure to compute reasonable losses at frequencies of 250-500 Hz for the 

 winter profile make this figure suspect. Since this sediment model (see Section 3.5) was 

 of interim status, further investigation of this type of computational profile will be made 

 at a future date. However, the optimum frequency of propagation for the positive gradient, 

 winter case (Figure 4. 1 3) is probably between 200 and 650 Hz. 



Figures 4.14 and 4.15 show propagation losses calculated for the winter and summer 

 profile conditions, respectively, and for a sea floor sediment model without the 3-m sand 

 layer. Optimum frequencies of propagation are about 475 Hz for winter conditions and 

 at low frequencies (less than 50 Hz) for the summer profile conditions. High bottom reflec- 

 tion losses and downward refraction combine to give very high losses for summer (Figure 

 4.1 5). A 100 m source depth was used to obtain losses at ranges of 25 and 50 km. Propa- 

 gation losses at 50 km are about 40 dB higher at optimum frequencies for the summer 

 profile versus the winter profile (1 14 dB at 40 Hz (Figure 4.15) versus 74 dB at 475 Hz 

 (Figure 4.14)). At 100 Hz the propagation loss is greater by over 60 dB at 50 km for the 

 summer profile. 



KOREA STRAIT 

 WINTER PROFILE 

 25 m SOURCE 



J ' ' ■ ' ' 



J L 



500 700 1 000 



FREQUENCY, Hz 



Figure 4.13. Propagation loss for optimum receiver depth vs frequency for ranges 

 of 50, 1 00 and 150 km. Korea Strait winter profile, source depth 25 m. 



50 



